Monday, March 23 - Vehicle Data
Agenda:
- review of TODOs
- /TODO/ all CCS participants to review the Geotab document on EV data access requirements
- /TODO/ TBD (Benjamin ?) to compare Geotab doc on EV data requirements and the content of the data bundles wiki page
- /TODO/ Rex to review the Adaptive Autosar signal-to-service specifications
Participants
15:00 CET call: Benjamin (W3C), Gunnar, Philippe
16:00 CET call: Keith, Kevin V, Guru, Philippe
Minutes
First call
Second call
- /TODO/ all CCS participants to review the Geotab document on EV data access requirements
- Philippe reminds this TODO to participants, Kevin will look at Geotab document this week
- /TODO/ Rex to review the Adaptive Autosar signal-to-service specifications
- Philippe asks Keith (who knows well the Adaptive Autosar stack) whether he could provide an overview of the signal-to-service specifications to the CCS team
- Keith: will do,
Jira ticket created for tracking
- sync on the communication framework
- Keith was able to join today after a long period of overlapping between CCS call and other calls that prevented him from joining
- Kevin: for the sake of synchronizing Keith with the CCS current status of work, Kevin presents the CCS PoC as described in CCS Proof-Of-Concept - Work Breakdown Structure
- Keith: asks how we can get the gps position of a fleet of cars using the communication infrastructure proposed in the poc
- Kevin V / Philippe: we might have different variants of the poc depending on whether we simulate the vehicle data or we use existing devices like the Geotab Go device if we want to connect to actual cars
- Philippe: for the vehicle simulation, two options have been discussed in the GENIVI AASIG project which is also working on a poc : either using the vehicle simulator published by JLR on GENIVI gitub (genivi-vehicle-simulator repository) or using opends
- Keith: indicates that LGE Labs in the Silicon Valley have also a vehicle simulator
- Philippe: although we will start with laptops as execution targets for the poc, we might switch to automotive boards for the in-vehicle segment of the poc in the future, FYI Renesas has set up a lava-based test farm (with R-Car boards) to build and run the GENIVI CI process, we could use the test farm to simulate the vehicle segment
- Keith: FYI Adaptive Autosar has decommissionned their test farm and runs the CI process on qemu only
- Philippe: invites Keith to the second call of the week which is dedicated to the communication infrastructure
Wednesday, March 18 - Communication Framework
Agenda:
Participants
Ted, Kevin V,, Kevin de Sousa, Ulf, Gerald, Stephen, Gunnar, Philippe
Minutes
Communication Framework PoC staffing & WBS update
- Philippe: reminds that it is now time for the team need to pick up some of work items for the CCS PoC implementation,
- Ulf: I can take over the data server, and its connection to the state storage in the in-vehicle OS box, I am also interested in the data lake
Kevin: we can provide sample applications for the 3rd party service and neutral server marketplace boxes, i.e. entities that will consume graphql apis
Philippe: what about the OEM cloud box ?
Gunnar: we need to determine the signals and the use cases we want to show
Kevin V: we should show different data categories including big data chunks
Kevin V: showing anonymized data is not really possible because this requires a lot of data, it would be better to show different categories
- short discussion on the demo use cases
- Philippe: explains the use cases selected by the AASIG for the PoC demo (EV data, speed & tyre monitoring and air conditioning
- need to add the use cases definition work item
Gerald: I might be able to pull in someone in the in-vehicle OS box , I will ask Sebastian and Christian whether they could recycle something from the kuksa project for the state storage
Gunnar: we need to agree on the format for storing values, e.g. time stamps and values of different types
Gunnar: I would like to get an example of a query for getting multiple snapshots of vehicle data in order to make graphql shine a little bit
Kevin V: if you look at the state of the vehicle, you have a benefit of using a graphql query (get multiple data in one query)
Kevin V: it should also possible to push data with graphql
discussion on the protocols between boxes
Gunnar: I am still waiting for an actual schema for using graphql from someone
Philippe: what about Geotab / Canada ?
Kevin S: I am still chasing what GENIVI is doing, I see myself in a support role rather
review and update of the owner column in the WBS table
Ulf: someone else has to take over the vehicle simulation
Philippe: execution targets will be laptops for the initial implementation
- Vehicle client
- vehicle client: Sanjeev's work does not seem to be directly reusable
- Ulf: the client would be repsonsible for stoting the data into the data lake
- Data lake
- Gunnar: concenring the data lake, do we have a database person in the call ?
- KevinV not on my side
- Ulf: this is where we should try to get BMW in, it won't be a good idea to invent our own database structure
- Ted: their solution is based on VSS/VSSo too, trying to get Adnan to present what he showed at last W3C F2F
- Gunnar: I talked to Adnan yesterda
- OEM cloud
Gunnar: what the resource management is all about in this box ?
Kevin V: resource management comes from the extended vehicle ISO standard
Gunnar: can we replace this naming by "data server apis" in the OEM cloud
Kevin V: agreed
Neutral Server market
Miscellaneous
- Test farm
Stephen: I checked the use of lava farm for running the CCS poc, do you want to use a RTOS in the in-vehicle OS box or linux ?
Gunnar: we can plan for this at a later time, The PoC will be linux-based including the in-vehicle OS box
Philippe: however Renesas could an Adaptive Autosar demonstrator stack in the in-vehicle OS box, this would be good for the demo, it will still be linux-based because the Adaptive Autosar demonstrator is based in RT Linuw
Monday, March 16 - Vehicle Data
Agenda:
- EV data access requirements
- data contracts & data categories
- Bosch Vehicle Abstraction Layer - relevance for CCS
- communication framework - update
- AOB
Participants
Ted W3C (guest), Sanjeev (Samsung), Benjamin (W3C), Kevin Valdek, Glenn, Kevin de Sousa, Ulf, Rex, Gunnar, Philippe
- apologies: Gerald
Minutes
EV data access requirements
- Glenn has published the EV data access requirements document as expected
- the document was shared with Ulf and might be introduced in W3C
- Glenn: underlines the use cases presented in the document
- /TODO/ all CCS participants to review the Geotab document on EV data access requirements
data contracts and data categories
- Gunnar: is looking at the splitting the Jira ticket
as expected
Bosch Vehicle Abstraction Layer - relevance for CCS
- Gunnar: summarizes the discussion with Bosch that happened last Wednesday
- Philippe: reminds the rationale for having a discussion with Bosch (which is organizing a workshop on vehicle abstraction at the AMM)
- Gunnar: there is a need to unify the industry around a so-called vehicle common interface model, I do not know whether we need to mix the data driven like VSS and the vehicle API
- ULf : VSS is good for data decription, service description needs to be added to it, there is also a need to analyze how to map the VSS signals (using the VSS terminology) to whathever is below, ie the different automotive buses
- Ulf: I know thar BMW and Volvo have thoughts about this
- Gunnar: would be happy to get the correct contacts at Volvo (VCC) to pursue this discussion with them
- Ulf: someip to VSS is exactly what VCC has done, Peter Winzell is the one to talk to, he might convince VCC to share some of the work they did
communication framework
- Gunnar: shows the CCS PoC WBS
- Gunnar: aks Sanjeev for clarification on the work he is doing on the so-called clients
- Sanjeev: I limited the scope of my work to the W3C demo shceduled un a couple of weeks time
- Sanjeev: I implemented http and websocket connections for the Tizen watch, we have now the vehicle dashboard on the watch, the java scripts used in this implementation are quite straitghforward but need to be rewritten for better clarity
- Sanjeev the watch code is in the app store, but I intenf to publish the code as well, the web interface is avauable is the daemon folder, the watch interface is pretty much a html interface
- Ted: Magnus from Melco set up the Gen2 server instance on MIT cloud host, the app could point to that
- Ted: has set up a Gen2 server instance for demo purposes
- Ulf: this data server is up and running, it is available but the link to it has not been released yet
- Ted: link is https://www.w3.org/auto/wg/wiki/Vss_data
- Sanjeev: back to the clients I am developing, I want to separate the library interfaces from the UX
- Gunnar: we need to figure out an idea of the demo we want to do, we need to major progress on this actually
- Kevin V: there is a lot of different use cases, that could be used to demonstrate the PoC
- Gunnar: we can work with simple queries, but it owuld be good to have simple queries moving a lot of data in odrer to learn abour possible bottlenecks in our communication framework proposal
- Benjamin: we could reuse what we have introduced in the data bundles
- Rex: do you have at Geotab a document listing all the telematics data ?
- Glenn: I am not aware of it
- /TODO/ TBD (Benjamin ?) to compare Geotab doc on EV data requirements and the content of the data bundles wiki page
- Ulf: asks whether Gunnar can present the CCS PoC block diagram in next week's W3C virtual meeting
- Gunnar: will do
graphql
- Gunnar: I need help from anyone who is familier with it
- Gunnar: we have a limited set of metadata in VSS, VSS is not following a graph and I ask people who proposed graphql what their vision is all about
- Benjamin; graphql in theory works with graphs, it does not perfom as well with trees and taxonomies
- Benjmain: in the case of VSS, you have VSS tree and you look at signals that are spread over the VSS tree, you define schemas for this purpose
- Benjamin: if you want a more complex relationship between signals, the computation you need will be much higher
- Benjamin: please have a look at https://graphql.org/learn/schema/, for querying and pushing data regularly graphql is appropritate
- Gunnar: Example ?
- discussioon on how to represent the vehicle data structure
- Kevin V: do you mean nested relationships ?
- Gunnar: yes for instance
- Gunnar: I would appreciate some hints about graphql
- Ted: Adnan could present some stuff on graphql related to vsso and the graph project, to be followed by Gunnar
AOB
- next week's W3C virtual workshop
Wednesday, March 11 - Communication Framework
Agenda:
- presentation of CCS PoC for Ted (W3C)
- AOB
Participants
Ted W3C (guest), Kevin de Sousa, Ulf, Gunnar, Philippe
- apologies: Kevin V
Minutes
Communication Framework
- Ulf: informed Ted (W3C) about the work we are planning to do on the CCS PoC and invited him to join the call
- Gunnar: introduces briefly the 2 main thrads of activities in the CCS project: vehicle data model and communication infrastructure, then presents the CCS PoC wiki page CCS Proof-Of-Concept - Work Breakdown Structure
- Ted: are you in contact with Graham Smethurs at BMW ?
- Philippe: Yes, there is an action item to talk to him
- Ted: are the CCS PoC aligned with BMW ?
- Gunnar: yes; we are very much aligned on the data server between GENIVI CCS and AASIG projects , and BMW is active in AASIG and promotes graphql there
- short discussion about Sanjeev who completed the Gen2 clients on Tizen (for the Samsung watch), Linux, and Android, Sanjeev seems to be looking at graphql as well (TBC)
- Philippe: I have an action item to contact Sanjeev that we happen to know very well due to former collaboration on demonstrators
- Ted: I talked to John S at Ford recently, Ford is still in the process of making plans concerning the vehicle data APIs
- Philippe: @Ted John delivered a talk in the CCS project track at the GENIVI tech summit of Nov 2019 in Troy, USA, slide deck is here
- Gunnar & Philippe invite Ted to join the CCS weekly calls
- Ted will be added to the CCS mailing list
Monday, March 9 - Vehicle Data
Agenda:
- data contracts & data categories
- Bosch Vehicle Abstraction Layer - relevance for CCS
- communication framework - short update
- Sprint & backlog review
- look at active sprint & backlog of CCS tracker
- AOB
Participants
Ulf, Glenn, Rex, Guru, Kevin Valdek, Gunnar, Philippe
Minutes
Vehicle data model enhancement
- Glenn: Geotab EV division drafted a paper containing the EV signals, e.g. charging information, discharge of the battery, etc. when combined with gps and timestamp this data set can solve all the use cases like range vs destination, upcoming load for the charging station, etc., should be published within some days
- call with Bosch happened on 11 March, debrief is at the agenda of Monday 16 March call
data contracts & data categories
- Gunnar: data categories and data contracts is rather a tool than a result, relates to access control, this topic is also being discussed in W3C
- Ulf: we are talking about data categorization and VSS layers but I cannot see the link between both
- Gunnar: there are 2 ways of representing the categories in the layer concept, VSS layer is a format to represent the data categories
- Ulf: discussion data catzgorization is different from representing the categories for instance using the VSS layer data format
- Gunnar: access control group is one example of data category (TBC)
- Gunnar: data contract is something different
- Ulf: it is a fuzzy concept for me
- Gunnar: I can see the connection, I need to check the Geotab's presentation in W3C team about it
- the presentation is available there
- Gunnar: we need to have common terms on contracts, technical and commercial parameters
- Rex: when you talk about QoS, do you have in mind protocols like DDS and Some/IP (signal-to-service), ARA::COM, some of the building blocks may be there
- Philippe: asks Rex to review the signal-to-service requirements in ARA:COM
created
/TODO/ Philippe send to Rex the Adaptive Autosar signal-to-service specification DONE- Ulf: who are the two parties in such a contract discussion ?
- Gunnar: could be the OEMs and the neutral server companies
- Kevin: data providers = OEMs, there is a contract between the neutral server and the third parties (data consumers), the neutral server does not change much the content
- Gunnar: we need a framework / template to write those contracts
- Gunnar: we need to set up a framework for access control
- Kevin: applying data categories to data contracts is the real pain
- Gunnar let separate those two wordings and check whether the industry wants to progress on both topics
- /TODO/ Gunnar structure better the Jira ticket in order to distinguish the work on data contracts from the work on data categories
- Glenn: data contracts in Geotab document mentioned above might not fully be appropriate for the discussion we have here
- discussion on what a contract is all about
Communication Framework PoC - WBS
- Kevin presents the wiki page containing the WBS
- /TODO/ all participants to review the WBS and identify which work item could be taken over
- Gunnar shows
where he added the list of technogolies that could be used for the PoC - Ulf: regarding the client implementation Sanjeev is doing, Sanjeev will present his work as part of the remote F2F W3C will have at the end of the month, Sanjeev is working on different platforms including an Android app
- OEM cloud: Kevin is fine with Hadoop
- Glenn: we need to remap the data from different OEMs to a common standard, go back to the data guys and talk to them about the pinpoints
- /TODO/ Glenn provide feedback on a possible common standard for next week
- /TODO/ Philippe set up a call with Sanjeev and invite him to follow / join the CCS project
- Rex: IMHO we could set an Adaptive Autosar stack in the cloud
- Philippe: possibly but in my opinion this is for the future, the architecture of the current PoC adresses a mid-term vehicle EE architecture (like the one depicted on slide #4 and marked as "today" of the VAP presentation by Bosch), having an Autosar stack in the cloud corresponfs rather to a future vehicle EE architecture like the one depicted on slide #4 and marked as "smart infrastructure" of the VAP presentation by Bosch
Wednesday, March 4 Communication Framework
Agenda:
- PoC implementation
- how to instantiate the big picture on different vehicle EE / Cloud architectures
- AOB
Participants
Gerald (partial), Ulf, Kevin Valdek, Eex (Saferide), Gunnar, Philippe
Minutes
PoC implementation
Ulf: would like to advertize the prototype Gen2 server implementation we do for the W3C group
Kevin: supports the proposal
Ulf: it is not production level but there is a considerable number of functionalities available
Ulf: error handling is not so resilient yet
Gunnar: agreed, we need to put this into the blue boxes of the big picture diagram, blue boxes named data server could be implemented using the Gen2 prototype
Gunnar: graphql is another option that could be demonstrated in a parallel thread
Philippe: VHAL team is looking at graphql implementation
Gunnar: this is an argumen tor CCS to look into Gen2
Gunnar: what about the clients ?
Ulf: we have small clients available for testing the data servers, Sanjeev (former Samsung) is working on clients
/TODO/ Philippe contact Sanjeev
Ulf: would be good to collaborate with the W3C project implementing the data lake
Phil: would be good to augment the big picture diagram (look at diagram V2 in Vehicle data exchange protocols) with the list of work items and technos used and create a table listing the wok items and the assignees like we did
to contributors
/TODO/ Kevin to add link to possible technologies on the diagram
Ulf: we can use for instance https://github.com/MEAE-GOT/W3C_VehicleSignalInterfaceImpl
/TODO/ Gunnar do the same
- updated
- Kevin: we need to determine what we use on the OEM cloud part
discussion on the use of Gen2
Gunnar: let determine who will provide the graphql implementation , maybe BMW maybe myself (I have to put some work into it)
discussion on how to align the CCS PoC project and the W3C Gen2 implementation
Gunnar: we all agree to align projects, what we need are resources
Rex: I have a common terminology question, which translation / transformation are you talking about ?
Ulf: BMW has a large number of cars connected today, they do not use VSS format, they use different proprietary data formats
Ulf: BMW took the decision that once these data are in the cloud, BMW will create the data lake using VSS1 or VSS2 and use graphql to do queries
Rex: wonders whether BMW is using the ASAM standard for describing signals
Rex: how is the topic of data compression approach ?
Gunnar: it could typically be done at the transport protocol level
Gunnar: what do you, Rex, mean by analyzing compressed data ? we have not touched this yet
Gunnar: do we need to start looking into OpenID ?
/TODO/ Kevin initiate the WBS table for the CCS PoC from the example provided here by the VHAL project
created as a sub-task of 
AMM preparation
- GENIVI is currently actively doing the AMM content planning, Steve Crumb (Executive Director) is watching closely the virus situation and a decision to keep the AMM physical vs virtual will happen soon
- the program includes a CCS track with sessions on communication framework and vehicle data model and PoC demo and likely a session called possibly Looking forward: towards a Common Vehicle Interface (TBC)
Monday, March 2 - Vehicle Data
Agenda:
- a use case presented by Gerald
- Data set for EV management
- data contracts & data categories
- communication framework
- Sprint & backlog review
- AOB
Participants
Ulf, Glenn, Gerald, Guru, Kevin Valdek, Gunnar, Philippe
Minutes
a use case presented by Gerald
Gerald shows this slide (Gerald Spreitz please upload the slide), is the CCS use case APP presented here still what we are discussing in the project ?
Gunnar: what do you mean by: is CCS still supporting this use case ?
- Gunnar: at signals level, we can think of signal groups you can get access to
- Ulf: W3C Gen2 could be a potential candidate for a large part of this use case, Gen2 supports both on-board agents and cloud-based agents
Gerald: can you clarify what you mean by the grouping of signals ?
Gunnar: tthis is work-in-progress, my view is that there will be many groups because having a fine granularity is key
Gerald: can the granularity be down to one signal only ?
Gunnar: I wrote an example related to vehicle identification for the VSS layer presentation that gives an idea of what I think is an appropriate grouping
Gunnar: we could look a little bit deeper in the design of this APP use case you presented using Gen2
created, unassigned
Gerald: thinks about the monetization of this use case, if you do not provide vehicle data you have to pay for the app
Gerald: we should not forget about the commercial aspects in our work
Data set for EV management
- Gunnar: remember that data set must rather be called naming of signals
- Glenn: this is work-in-progress, expects to have an answer in a week or the rationale for not having an answer and not being able to share the data Geotab use with power utilities
- Glenn: what is important is the rationale of the power utilities for using the signals
- commented
data contracts & data categories
- Gunnar: shows the data bundles wiki page, reminds about the privacy aspects
- Gunnar: the need has to drive the work, we need to work out a complete access control group first before reworking data contracts and data categories
- Gunnar: do other think we need to work in parallel ?
- Gunnar: elaborates on the EV data the OEMs would like to control the accessibility of (e.g. battery durability relales to warranty cost, etc.), I am sure we will have the mechanisms to handle this
- the data related to this might even not be transmitted because providing those data is not in the business contract
- delivering that data can be done under a contract to specific third parties
- Gunnar: we expect to have the mechanism to control the access technically and businesswise
created, unassigned
communication framework
- look at Vehicle Data Exchange Prococols wiki page
- Gunnar: presented the wiki page, are the protocols we discuss enough for what we want to do ? he explains his vision of the work to do as described in the wiki
- Kevin: explains the differences between the latest "big picture" diagram and the initial one
- Kevin: explains he removed the container processing because we use Gen2 now
- Gunnar: here are in my the pieces of the puzzle
- understanding of the big picture
- protocols between compoments
- dig into components
Sprint & backlog review
- the Jira tracker was renamed as CCS (instead of VCS) and updated, look at CCS backlog
- /TODO/ Gunnar to check that the Jira notification scheme is active for the CCS project
- /TODO/ All to check the backlog and current sprint content
- current sprint is until the end of March
- Philippe's expectation is that we will be able to start some PoC activities soon in order to show something at the upcoming Spring AMM (12-14 May)
AOB Privacy
- Glenn: the following topic might be of interest to the team _ Privacy and EU
- Do you know if the organization (or any of the OEMs) are planning to weigh-in on the following ==> The European Data Protection Board (EDPB) releases DRAFT guidelines for processing of Personal Data in the context of Connected Vehicles - and is open for comment --> Guidelines 1/2020 (Comments Due on March 20th)
- links
- /TODO/ Gerald to review the guidelines and provide a summary to the team
- Philippe (added offline): IMHO we need to forward Glenn's question both to the OEMs and the GENIVI security team
, 
created
Friday, February 28 - Communication Framework
Agenda:
- continuation of discussion started in Monday 24 February call (W3C)
- how to instantiate the big picture on different vehicle EE / Cloud architectures
- AOB
Participants
Gerald, Ulf, Kevin de Sousa (partially joined), Kevin Valdek, Gunnar, Philippe
Minutes
Communication Framework - W3C
Kevin checked Gen2
how Gen2 fits in the big picture: both rest and web sockets
explanation of what the CVIM container processing is all about (some kind of buffering), what happens if the connection is lost
Ulf: this is a good point, there is no buffering in the standard W3C work
Ulf: this requiement has not been put on the table in the W3C team, maybe we should raise the point in W3C, what do you think about it, Gunnar ?
Gunnar: I an more on the solution mode than the standardization side, the container holds the measured values, there are a couple of design options
you can build a function in the protocol itself and request the sequence of data points for the last hour for instance
you can also reuse VSS to represent a sequence of signals
you can also consider a pre-processing of the signals like an histogram in the car, i.e. on the edge, there is some metadata to be transfered to achieve this
Gunnar: we need to identify the boundaries of the W3C protocols and how we can complement / augment this work
Kevin: in today's implementation, the signals are lost (no memory of the sequence of signals), if we assume there is a value storage in the vehicle, we could perhaps send a request for giving the rpm for the last 3 hours
discussion on the data update process, ask for instance for the min/max data for the last couple of hours
Ulf: there is a possibility to use the query we have introduced in Gen2 for such a purpose
Ulf: go the filtering section (gencore.hmtl), you can search more than one node, you can also add metadata, in this context, it is possible to add some functionalities like the one we discussed before (with a time dimension)
Gunnar: it is possible for an OEM to override the metadata of a signal
Next step
- Gunnar: we have some work to do to design the functions that we want: min/max, histograms, etc. and look how we can implement these, either in the query language or in the metadata
- Gunnar: we need to gather the requirements ar the feature level
- Ulf: Geotab has a pattern / method we use for "friends" to minimize the data we keep, the computation draws straight lines between points, we use this method for all data we send from the embedded devices we use
- Gunnar: it sounds that we need a new wiki page describing the data processing features needed
/TODO/ Gunnar to initiate the wiki page DONE: Vehicle data exchange protocols- /TODO/ Kevin to add an update version of the big picture diagram and the description of some components
created for tracking
/TODO/ Philippe to send an invite for the Wednesday call DONE
Monday, February 24 - Vehicle Data
Agenda:
- Data set for EV management
- Communication Framework
- Vehicle Abstraction Layer
- CCS project backlog in Jira
- AOB
Participants
Ulf, Glenn, Kevin de Sousa (Geotab), Gerald (Bosch), Kevin Valdek (High-Mobility), Keith (LGE), Gunnar, Philippe
Minutes
Data set for EV management
- KevinV: the data set is not complete and some parties do not want to share all details
- Gunnar: You mean the OEM not wanting to share some of the detailed EV data
- Gunnar: Note that we discussed separating the standardization from the obligation to deliver the data.
- The standards _enable_ data sharing between partners.
- Agreeing on actually delivering the data is more of a business contract.
- We can work on a business framework to enable this
- Gerald: Are there any standards for the agreement on which data is delivered ?
- Kevin: Yes some consent discussions exist but standards are not quite there yet, implementation of connected EVs is not widespread
- Gerald: But do proposals exist ?
- Kevin: Only what is mentioned in Extended Vehicle, to my knowledge (not a lot of details)
- Gerald: So is this an open point to study, for CCS project?
- Gunnar: Yes it could be but for me some of those places where this might apply will naturally "fall out" of further work on the architecture (meaning let's progress the technical architecture a bit
more and maybe later the opportunity to fill those gaps come up) - Kevin: (A more important) missing thing is consensus on what type of data is required for (thid party) features, what actually needs to be shared, to enable a particular service.
- Gunnar: So it ties into data categories and also access control groups might give some of that picture.
- Kevin: Agreed.
- Glenn: (So it's about) data requirements per use case. EV is a good one to start with because for other vehicles connectivity is a nice to have. For EVs it is critical, i.e. know where charging exists, and how/where vehicles exist and where charging is expected to happen (demand for capacity)
- Glenn: Therefore keep progressing the minimal set of data for EVs first.
- Next (added offline by Philippe): there is a need to complement / rework the description of the work items on data contracts and data categories in Jira
- look at
Discussion on vehicle abstraction layer.
- Several people missed this presentation last week.
- Gunnar: I at least need another round with Sebastian to fully grasp what we really mean here.
- Gunnar: In my opinion we should emphasize commonality, not differences.
- Work towards that common interfaces and technology should be used throughout the industry, rather than emphasize that common interfaces are needed / because / common technology / standards are NOT used throughout
- I might be missing details but I'm not really comfortable with the "abstraction" idea proposed by Sebastian
- Next (added offline by Philippe): there is a need to complement / rework the description of the related work items in Jira
- look at
Communication framework
- the high-level block diagram proposed by KevinV is shown
- KevinS: brings a clarification on what Geotab is doing today, Geotab runs a sort of container on their embedded device today
- KevinS: for an OEM, it might consist in running a GENIVI VSS as a so-called container
- KevinS: what Geotab is doing is generalizing the in-vehicle processing with the OEM cloud
- KevinS: the communication between the Geotab device and the cloud is proprierary
- KevinS: the Geotab communication framework architecture currently is very closed to what is shown on the block diagram, the data processing would be very different on the Getoab side
- Ulf: the diagram is quite interesting, but where are the W3C standards mentioned? We see W3C Gen 2 to be also used between cloud and the vehicle
- Ulf: since Gunnar is active in W3C and we are trying to standardize the cloud to vehicle interaction, it would be worth showing this to W3C and discussing the transport protocol and the container processing
- Philippe: explains the approach taken by KevinS to build to this diagram
- Ulf: VISS is the first assumption for the implementaIion, we are now working with the so-called Gen2 in W3C
- Gunnar: Let's see if for every place where it says "REST", we can clarify that we mean "W3C Gen 2 REST", (sync with Kevin V to see if this is the case)
- /TODO/ Kevin V to look at Gen2
/TODO/ Ulf to send the links to Gen2 DONE below are some links to the W3C automotive working group.- Keith: what is VSS 2.0 about ?
- Gunnar: it only just means "next release". It is not a new concept. For now consider the master branch as the content. Sure, there have been some advancement of the syntax, semantics, tooling and vss-layers, but the "2" is not there to indicate a whole new concept.
- Gunnar: clarification brought on how to bring the latest values, VSS define only the isgnals, we need data feeders
- Gunnar: querying the data and ask for their values is something that W3C is not looking into yet, the CCS project is going beyond what W3C is doing
- KevinV: graphQL is another Web technology; BMW has an interest in it in addition to what Google is doing, explains the thinking behind bringing grahQL in the picture
- Ulf: I agree that (the powerful query language of) graphQL is very powerful, it is a perfect technology to work with at the data lake level, however I have some doubt about that approach if you put it in the vehicle,
- Ulf: The interpretation and execution of the queries in the vehicle will put equally a powerful load on the vehicle to answer those
- Philippe: explains the short term to long term of the vehicle EE (look at slide #4 of the EE architecture evolution, in the long-term, we might afford grapQL on in-vehicle HPC nodes (Hogh Power Compute).
- Philippe: The powerful computing nodes in the vehicle might be able to handle it.
- discussion on how to instantiate the CCS block-diagram on vehicle EE architectures. Geotab approach with an embedded device might be appropriate for the short to mid term. However CCS architecture draft proposal for the CCS block-diagram should rather be mapped on an architecture like the one of slide #4 above
- discussion on communication framework architecture will continue on Wednesday call at 5pm CET.
- A calendar invite has been sent
CCS project backlog in Jira
Monday, February 17 - Vehicle Data
Agenda:
- Data set for EV management
- Kuksa project update / Vehicle Abstraction Layer
- AOB
Participants
Ulf (Geotab), Sebastian (Bosch), Gerald, Guru, Gunnar, Philippe
- apologies: Glenn (bank holiday in Canada)
Minutes
Data set for EV management
- Ulf joins today's call following a request made by Glenn, Ulf is in CET timezone, Ulf checked the VSS tree for data related to EVs
- Ulf works rather with the structure of the tree and not on the content
- the tree contains many branches, I did a grep on EV data and found only one file called battery management containing data like temperature, charge status, there is also an charging inlet which is an enum that contains values like AC, DC
- in the drivetrain branch , there is an electric motor branch but with no data in it
- Philippe: who is in the position to complement the VSS tree ?
- Gunnar: complementing the tree is not a W3C process question , we need to spot the identify domain expert to fill the blanks
- Ulf: I agree with Gunnar, we need domain experts from OEMs to get the EV branch populated, I asked Volvo (VCC) before (when I was working with them), and got very small success
- Gunnar: this needs to get into the tree in order to get things move on with EVs
- Gunnar: we will follow up with Sarah (see minutes of CCS call of 10 February) , we can use whatever is available through Geotab devices (connected to OBD port today), since we do not know what is available from the OEMs yet
- Gunnar: a key question with the battery data is that OEMs might not be willing to share the data (e.g. battery capacity) that would be related to battery degradation
- Gunnar: it would be good to check which diagnostic data on battery we could get
- Philippe: what about W3C ?
- Gunnar: Ted is willing to get inputs on EV, but our interest is to get this done faster
- Ulf: if OEMs are reluctant to bring inputs, we might initiate this gathering by asking Geotab for the data they are used to
- Philippe: I will follow-up with Sarah
Kuksa project update / Vehicle Abstraction Layer
- Sebastian presents one slide giving the main timeline of the Kuksa projet and the follow-up project called VAL
- Appstacle project is over
- VAPP is an internal Bosch project, we are allowed to share on it, VAL stands for Vehicle Abstraction Layer
- Eclipse KUKSA.val project contains the Kuksa VISS server that was spinned off as a separate component
- Then Sebastian shows a slide deck presenting the Vehicle Asbtraction Layer concept Bosch is now digging into
- the slide deck contains the well known Bosch slide on vehicle EE architecture evolution (slide #3) and the next step which is the EE architecture extension to cloud connectivity (slide #4)
- the end2end vehicle application architecture (from back end / cloud services) to domain and zone controller is shown (slide #7)
- a so-called Vehicle Application Plugin Platform is identified that would support the execution of Vehicle Services plugins that run on HPC (High Power Computer) nodes and not and deeply embedded microcontrollers
- one example of a Vehicle Service is for instance the the trunk opening (for putting packages in the trunk), the actual implementation is very different among OEMS, having a trunk opening service abstract description would help (look at slides #8-9-10)
- Gunnar: is VSS on the slides ?
- Sebastian: VSS was not there because these slides were a teaser about the VAL project (before kick-off)
- Ulf: will this work suitable for future VSS work ?
- Sebastian: not only one technology is suitable for data abstraction, VSS is one (look at slide #11)
- Philippe: do we need more abstraction beyond the trunk lock and trunk electric drive ?
- Gunnar: the trunk opening abstraction is something like a programming API
- Sebastian: the RPC is an easy question, w.r.t. abstraction, the trunk might be different for a car, a bus or a truck (this last one being very different)
- Sebastian: we can play with the Kuksa implementation of VSS, it is good for playing with the abstraction concepts
- Sebastian: however we do not intend to push the work we did in ISO (TBC)
- Gunnar: I agree that the current state of VSS is not enough to describe a vehicle, the goal in my opinion is that at the end of the road the signals in VSS are actually this
vehicle abstraction layer - Gunnar: the part / model / layer about the trunk opening is a (parallel) project different from VSS in my opinion
- Sebastian: agreed, the VAPP stuff is different from VSS
- Gunnar: I have no problem with the Bosch big picture and having more programming APIs
- Philippe: do you intend to describe the vehicle hardware elements as well ?
- Sebastian: we have signals and RPCs but we do not intend to describe the vehicle hw
- Gunnar: AFAIK Philippe refers to the vehicle configuration, e.g. in infotainment how many speakers do we have ?
- Gunnar: this configuration might be something to add to the VAPP description, what do you think ?
- Sebastian: my opinion is that the hw description is given by some sort of tooling whose role is to describe the "ugly part of the system", the VSS fits well for describing the abstraction
we need - Ulf: I agree that the stuff below should be abstracted, IMHO the tree structure used by VSS is also suitable for describing the "ugly stuff below", a few OEMs are using
the VSS tree to link things together, there is the layer mechanism we are using in W3C to organize/abstract the data that helps this work, but it is OEM specific work that can be
added to the framework - Gunnar: this is the proprietary data base that could be added using the same framework
- Gunnar: it is very unlikely that OEMs will redefine their CAN signals because of VSS, we will need a translation for those signals, however when we move to Ethernet, we could use VSS to describe the data (PDUs) on the Ethernet networks
- Sebastian: so called feeders will push the available data to the live tree representing the car wherever they come from (OBD, CAN, Ethernet, etc.)
- Gunnar: IMHO the live tree you mentioned is the data storage used by the vehicle data server
- Gunnar: today's discussion is helpful, we are synchronizing our vision of the vehicle abstraction
Review of TODOs
TODO Glenn review the current EV branch of the VSS database and check whether the data described there are relevant and sufficient for enabling EV charging management / DERMS WIP- Ulf provided his feedback on VSS today (on behalf of Glenn)
- Glenn check to verify that the EV data to satisfy these EV use cases are available in Sarah's presentation
TODO Philippe to get back to Sarah to get possible inputs from Geotab on the data used for EV management DONE- Sarah provided the link to her presentation on commercial EV management
Wednesday, February 12 - Communication Framework
Agenda:
- Communication framework big picture
- Next steps
- Review of TODOs
- AOB
Participants
Minutes
Communication Framework
Next steps
- Philippe: IMHO we need now to derive the scope of proof-of-concept work for the communication framework, we might consider organizing a F2F workshop (or a longer telco) for defining the work breakdown structure for this work
- KevinV: my company could possibly provide a contribution on the top left part of the framework ("neutral server marketplace" on the diagram)
- KevinV: IMHO Geotab could possibly provide inputs/code for the bottom right part of the framework ("container processing" on the diagram)
Next call
- next CCS - Communication Framework call will be scheduled on Wednesday 19 February at 5pm CET
- however we might come back to a single CCS weekly call on Mondays later.
Review of TODOs
/TODO/ Glenn to provide the link to the YouTube video presenting an example of data processing / compression on the vehicle side
/TODO/ KevinV to draw a rough sketch of of the end-to-end architecture and provide a big picture of the Vehicle2Cloud Communication Framework DONE
/TODO/ Gunnar to provide the link to the 5GAA paper describing the data foreseen in V2V/V2X data exchanges
/TODO/ Glenn to provide the links to NHTSA paper on V2V/V2X data exchanges
Monday, February 10 - Vehicle Data
Agenda:
- EV data use cases : presentation by Geotab & discussion
- communication infrastructure call schedule
- backlog review (skipped)
- AOB
Participants
Kevin, Glenn, Sarah, Benjamin (partial), Gunnar, Philippe
- apologies: Gerald, Guru, Keith
Minutes
EV data use cases : presentation by Geotab & discussion
- Sarah is senior customer projects manager in EV fleet management at Geotab and works on smart charging
- Geotab has a B2B & B2C smart charging management platform offering
- Sarah delivers this presentation
- The rationale for enabling smart charging through the monitoring of EV data is that the grid is not originally designed for EV charging and commercial EV fleet operators and power utilities expect to improve the charging operations according to different criteria (hence the different use cases presented)
- the consumer use case scenarii (i.e. for passenger EVs) are not distant from the commercial fleet, however there is less maturity in consumer EVs than commercial EVs
- DERMS means Distributed Energy Resource Management System, Demand-Side Management is a subset of DERMS
- Gunnar: we need to define the vehicle data we need for EV management in the VSS database we use in GENIVI for the vehicle data modeling
- Sarah: currently this is no standard wean use and it is a lot of work to do (i.e. to adapt to the commercial EVs our customer use), we are very much interested in standards
- Gunnar: you use the OBD-II port, it is a subset of all electrical data available in the car, in VSS the vehicle data set is much larger, the branch for EV data is very minimal
- Philippe: do you support OCPP (Open Charge Alliance) ?
- Sarah: we support OCPP in our smart charging platform, OCCP is not using a lot of vehicle data, we look at the state of charge in the vehicle only
- Philippe: explains GENIVI objective which is to enhance the VSS database w.r.t. EVs in order to be future proof (because ICE part will decrease in the future)
- Gunnar: this is why we need domain expertise to identify which data need to be added in VSS for EVs
- Sarah: it is not a considerable data set, OEMs have a different level of data than the one we use, fleet managers need less data, the so-called last mile delivery use case is interested in other set of
data, power utilities need only the 15mn average power consumption - Glenn: it is worth noting that with commercial fleet management, the decision making / policy making are focused in the same hands
- Glenn: in the consumer domain instead, the decision making is much more distributed, the policies are much more diverse, deploying DERMS is more difficult (or less mature for the time being)
- Next step
- Gunnar: we need to look at the current content of VSS and see how it matches with Geotab undertanding of the use cases and required data set
- Kevin: the datasets for EV are very limited because the OEMs do not want to share the information about their battery management (and degradation for instance)
- Glenn: we are in a process of mapping the vehicle data we use at Geotab artefacts to the VSS data but the project is not completed yet
- Gunnar: we need to improve the VSS database if necessary, e.g. to allow third parties to have access to some specific dataset extensions
- TODO Glenn review the current EV branch of the VSS database and check whether the data described there are relevant and sufficient for enabling EV charging management / DERMS
Communication infrastructure call schedule
- Next call will on Wednesday 12 February at 5pm CET. A calendar invite will be sent
Thursday, January 30 - Communication Framework
Agenda:
- brainstorming on Vehicle2Cloud Communication Framework
- AOB
Participants
- Kevin de Sousa, Glenn Atkinson (Geotab)
- Kevin Valdek (High Mobility)
- Gunnar, Philippe
Minutes
Goal setting for the call
- Gunnar: which protocols are relevant ? Web protocols (REST), MQTT, NiFi, etc.
- Gunnar: what is the rate for transferring the data ? The framework deal with things like when a piece of data needs to be requested in order to meet the caching freshness goals and when the data flow needs to slow down because a certain part in the flow is getting its buffers filled, caching data when the dataflow goes down (management of the back pressure on the protocols) ?
- KevinV: this is a good starting point
- KevinV: high frequency data cannot likely be supported by REST, because of required refreshing rate, this should be considered in the design of the architecture
- Gunnar: we need to mention edge computing because for some data it is necessary to transform the high volume of data, the data can be pre-processed on the edge (vehicle)
- KevinV: it is even the most complicated part, how are the data captured and processed ? can we influence this ?
- KevinS: everything sounds good, I am comfortable with the discussion so far and can bring experience with other oems in Northern America, there are pros and cons of the different approaches, will a standardized approach be based on edge computing ?
- Glenn: Gunnar brought the idea of having a buffer in the vehicle containing some data that can bring to the cloud some time later, going through dark spots of connectivity
- KevinS: this is the way Geotab devices are working today, we do memory buffers for the different data we are sampling
- Gunnar: AFAIK some data compression can be done by matching data to particular curves (splines)
- Glenn: the techno used by Geotab is presented in a 10mn video available on YouTube, I could certainly investigate whether we could push this to W3C
- /TODO/ Glenn to provide the link to the YouTube video
- KevinS: yes, it is appropriate
- KevinV: in the GENIVI CCS project we want to define a framework and if the inputs you just mentioned are relevant, we want the CCS framework to meet this kind of requirements, we should use VSS as well, we would request Geotab to tell the CCS project whether the features identified for the framework are valid
Deliverables
- KevinV: what could be the project deliverables ?
- Philippe: I agree this is important
- Gunnar: we need to describe a technical architecture and the software components in this architecture
- Gunnar: it is unknown whether we will get a consensus on the architecture among the industry
- Gunnar: we should list all the different users of data and the categories for those data and look at how combining them with the vehicle data
- Gunnar: as far as a deliverable is concerned, we could write a white paper, we need to determine how complete we can make such a paper
- KevinV: for categories, we have some materials, it would be good to have some technical architecture in place for supporting those categories
Backlog building
- KevinV: I can do a rough sketch of the end-to-end architecture and provide a big picture in 2-week time
- /TODO/ KevinV draw a rough sketch of of the end-to-end architecture and provide a big picture of the Vehicle2Cloud Communication Framework
- Glenn: I like the idea of a whitepaper, Geotab could draft a section of in-vehicle data capturing and processing
- Gleen: KevinS could provide some high-level slides on this as well
- Gunnar: we need to identify and list the different categories of data users and the rationale for using these data
- Gunnar: which data (like environment data) could be merged with vehicle data ?
- Glenn: we should take into account the DSRC-based data at intersections, the objective with these data is identify the risk associated with intersections
- Gunnar. We need to identify which data are foreseen in V2V/V2X exchanges, I have in mind a paper written by the 5GAA consortium
- /TODO/ Gunnar to provide the link to the 5GAA paper describing the data foreseen in V2V/V2X data exchanges, there are several papers on https://5gaa.org
- Glenn: NHTSA in the US has projects related to this, they have not decided whether the industry will be going be DSRC or 5G
- /TODO/ Glenn to provide the links to NHTSA paper on V2V/V2X data exchanges
Next call
- It will be likely scheduled on Wed 12 February at 5pm CET provided neither Gunnar and Philippe have an agenda collision in the same timeslot. This will be confirmed
Monday, January 27
Agenda:
- CES debriefing
- vehicle data models: relevance of the specs for EV Charging to our work on data modeling (e.g. OCPP)
- platform candidates for the CCS PoC: are the following platforms relevant : Amazon Auto, MS Azur, Veniam ?
- project organization: add another weekly call to discuss architecture
- Sprint & backlog review
- AOB
Participants
Kevin, Guru, Glenn, Benjamin, Keith, Gunnar, Philippe, Steve
Minutes
EV charging
- Kevin: IMHO EV charging when the vehicle is connected to the charging station is totally in the scope of use cases and specifications defined by a consortium like the Open Charging Alliance. However there is a need to know the status of the EV battery charge when the vehicle is not connected
- Glenn: AKAIK the utilities request the status of EV battery charge for load balancing, I could bring an expert in the call to explain the EV/Power Utility communication use case for commercial vehicles as means possible inclusion by GENIVI in defining 'scope' for EV/Electrical Utility communication requirements - leading to the development of open specifications for broaded use of same.
- /TODO/ Geotab to provide an expert in this area to describe current use cases at Monday GENIVI call. (5:30 CET)
- Benjamin: in W3C VSS, there is a branch on battery capacity but it is empty up to now
- Gunnar: agreed, there is a need to complement VSS on this
- Gunnar: I agree that when the EV is connected to the charging station, this is not a use case for GENIVI, however when the vehicle is OTA, this is in scope of our CCS project
- Kevin: last week we at High Mobility open sourced all of our SDK with a unified set of APIs (based on data we got from BMW/Mercedes/... on the production projects), it might be good to look at data examples available at
- Gunnar: IMHO we should rename this topic to EV management (i.e. something different than EV charging) since we do not overlap with the work on EV charging
- Guru: we need also to clarify how this topic fits with the EV battery management system
- EV charging docs (reminder) - useful links
Communication infrastructure - Platform candidates for the CCS PoC
- Philippe: reminds us it would be good to look at Amazon Auto and MS Azure and Veniam (https://veniam.com/products/) as possible target platforms for the CCS project demonstrator
Kevin: does not know Veniam, will check
Gunnar: we need to undestand what role Amazon would like to play: do Amazon want to compete with companies like Geotab and others or do they simply want to sell infrastructure
Gunnar: IMHO it is time to make a generalization of the infrastructure we have in mind
Gunnar: I would like to know whether the car OEMs are already communicating their data through vehicle neutral servers ?
Kevin: we should organize a call to brainstorm on the communication with others
Gunnar: AFAIK W3C process is a little bit slow, we should organize to make faster progress, we need to put building blocks in place
decision schedule a call for brainstorming on communication infrastructure with Kevin/Gunnar on Thu 30 Jan 11:30am CET (ad-hoc call), Philippe will send a calendar invite DONE
- (added offline): Geotab proposed to provide an overview of commercial vehicle to cloud services use case as an example that GENIVI may choose to include in TBD scope for vehicle/'cloud services' (e.g. Amazon, Microsoft Azure), and development of applicable open specifications to satisfy such use cases. Geotab will join the ad-hoc call to present this.
Data categories
Gunnar: in order to keep things concrete, it would be good to give an example of data categories for VSS signals
Glenn asks for clarification
Glenn will send a note to Geotab/Ulf and asks him to sync with Gunnar on this
Gunnar: the scope of the work proposed is not the same as what Geotab is doing for the W3C VSS PoC but we could reuse the same set of data
/TODO/ Gunnar follow with Ulf/Geotab on providing examples of data categories for VSS
Keith: aks a question on VISS (not captured, apologies)
Gunnar: will provide a clarification
Project organization
- doodle-like table to gather participants' availability for a second call updated online, table is here
- Wednesdays at 4pm CET or 5pm CET are the possible common slots
- /TODO/ Glenn to check whether these timeslots fit with the availabilities for Geotab
- /TO/ Guru to check whether these timeslots fit with his availabilities
- we will make the decision for the timeslot and the periodicity of calls at next Monday call
Sprint & backlog review
- Philippe: will update the backlog and the current sprint content according to latest discussions
- Glenn reports on the event he attended in Washington DC last week, this event organized by SAE was about the real-world data collection, 1600 people attended it, including many OEMs; Glenn mentioned GENIVI CSS work in the talk he delivered at the event
- Philippe: reminds us to identify other events in different geographies
AOB
- Due to traveling constraints, Philippe proposed to re-schedule next Monday call at 5:30pm CET
- All: agreed. A calendar invite will be sent.
Monday, January 20
Agenda:
- CES debriefing
- vehicle data models: relevance of the specs for EV Charging to our work on data modeling (e.g. OCPP)
- platform candidates for the CCS PoC: are the following platforms relevant : Amazon Auto, MS Azur, Veniam ?
- project organization: add another weekly call to discuss architecture
- Sprint & backlog review (skipped)
- AOB
Participants
Minutes
CES debrief
- GENIVI showcase: 1400 people attended it, Gunnar & Philippe stood up in front of the GENIVI projects table, slide decks presenting the deliverables recently published were shown on large TV monitors (automotive virtualization platform specifications, vehicle data models overview, FARACON tool). We had discussion with Microsoft, GM and Ford. Very good feedback from the participants
- SAE/GENIVI Connect2Car track: several panels were organized, discussion with Exelon Corp (a US power utility) about creating a WG in GENIVI on vehicle electrification
- first step would be for Exelon to attend the CCS call to discuss the need for vehicle data modeling for EV charging
- Transportation Technology track: Philippe attended a session called IoT and AI Revolutionizing Transportation, participants were Veniam CEO (https://veniam.com), VW Automony CEO (VW subsidiary dedicated to Autonomous Vehicle technology), Dell Technologies and a lawyer from the UK&Wales transportation commission (lawyer was specialized on AV deployment). Veniam CEO presented the 5G-based IOT platform they market to support the vehicle connectivity
- Automotive Exhibition Hall: among many other companies, Amazon Auto had a big booth there and showed many demos, Amazon Auto booth was crowded
EV charging
- docs to look at are the following
- Glenn: power utilities use history of EVs battery charge for balancing the power to supply EVs
- Philippe: IMHO it would be good to look at the public specifications for EV charging protocols and data contents and check for their relevance for the work on vehicle data modeling we have done
- to be further discussed next week
Platform candidates for the CCS PoC
- Philippe: IMHO it would be good to look at Amazon Auto and MS Azur and Veniam (https://veniam.com/products/) as possible target platforms for the CCS project demonstrator
- links pointing to Amazon Auto work
- GENIVI participant from Renesas said Renesas have done various integration projects with Amazon Auto and MS Azur.
- Gunnar: Iin my opinion, it is worth looking at these platforms
- to be further discussed next week
project organization
- Gerald suggested (offline) to have another call for the CCS project to discuss communication infrastructure architecture
- Guru: we need to agree on a one-week period or 2-week period for this additional call
- Philippe: will prepare a doodle-like table to gather participants' availability on a weekly basis, DONE table is here
AOB
- Philippe: we will have a AASIG Vehicle HAL F2F on 4-5 February at BMW, Munich
- Gunnar: started to discuss VSS layers on the W3C mailing list anf got a good feedback, VSS layers will be at the agenda of the AASIG Vehicle HAL F2F, link to VSS layers slide deck is in the minutes of the previous call
Monday, 16 December
Agenda:
- Dissemination: gap analysis deliverable - quick check on publishing
- Data categories - discussion on VSS layers and data categories
- Sprint & backlog review - Spring 6 content
Participants
Guru, Gerald, Kevin, Keith, Benjamin, Gunnar, Philippe
Minutes
Dissemination
- Philippe: gap analysis (actual title is Vehicle Data Models Overview and Gap Analysis) will be published this week together with the GENIVI newsletter
- then the CCS first deliverable will be advertized at GENIVI showcase at CES2020 together with other results, e.g. the Automotive Virtualization Platform Specification and the FARACON tool
- Philippe: reminds the participants to identify events, conferences, projects, forums where we could advertize the CCS work
Data categories - VSS layers :
Sprint & backlog review
- Philippe: shows the current sprint (Sprint 6) content - VCS tracker
done
it might be good to get back to Cloudera and have a chat with them possibly at CES 2020
Gerald indicated that the Kuksa could be able to deliver an update in January, Gerald will propose them to join one of the early Q1 calls - 27 January or 3 February
Gerald needs to re-check this topic, look at CCS minutes of Monday, September 23
- Philippe: IMHO we need to gather and synthesize all the materials / thoughts we had during Q4, 2019 and produce an updated vision of the end-to-end vehicle data capturing infrastructure
- Keith: is willing to engage into this but needs to do his ramp up first
- Kevin: would like to contribute also, needs more thinking
- Benjamin will check the business cases identified by W3C
created - Gerald will review the existing CCS materials and spot business cases identified in there
created - Gunnar: reminds that Glenn underlined that some aggregated data might become (privacy) sensitive while the elementary data contributing to the aggregation are not
AOB
- next call is scheduled on January 20, 4pm CET
- Enjoy the break !
Monday, December 9
Participants
Guru,, Gerald, Glenn, Kevin, Gunnar, Philippe, Steve
- apologies: Benjamin, Keith
Minutes
Review of the deliverable:
Data categories / "VSS layers"
- Gunnar introduces the draft slide deck he prepared following discussions with Benjamin and Kevin performed since last week's call
- this document introduces the concept of VSS layers which is a first step towards the communication framework design
- /TODO/ all review the slide deck on VSS layers
- in next week's call, Gunnar will go through the slide deck and answers questions.
Jira tracker
- /TODO/ "it would be good" to review the work items in Jira - link: VCS board
Monday, December 2
Participants
Keith, Guru,, Gerald, Kevin, Gunnar, Philippe, Steve
- apologies: Benjmain
Minutes
Review of the deliverable:
- document updated online
- Gunnar: inputs on Vehicle HAL still WIP (Gunnar)
- Philippe: our objective is to finish the polishing by the end of the week and hand it over to Steve/Traci for post-editing and publishing
- Kevin: some info on Android Automotive stakeholders is provided there
Communication framework design
- Gerald asks about the status of this thread
- Philippe: we need to build on the materials presented by Kevin in the related session of the tech summit, look here
- Philippe: IMHO our objective should be to refine the end-to-end architecture coined by Gerald at the beginning of the project and determine whether we have variants of the architecture depending on the use cases, we need also to detail further how the access to vehicle data is handled depending on various access policies
- Gunnar explains that categorizing the vehicle data is an important preliminary step for the design work
- /TODO/ Philippe & Gunnar organize a call with Benjamin on data categorization (on top of VSS) DONE
- the call with Benjamin was based on the following slide deck
- data categorization will be at the agenda of next week's call
Monday, November 25
Participants
Glenn, Keith, Benjamin, Gerald, Gunnar, Philippe
- apologies: Guru, Steve
Minutes
Review of the deliverable:
- document updated online
- need to add a table comparing the data models per criterion similar to the tables shown by Benjamin at the tech summit on the presentation on data models
- /TODO/ Benjamin add the table(s) comparing the data models to the Google docs document
- Glenn: what is the meaning of gap ? is the gap between protocols or between the data and the use cases
- Gunnar: the gap we have been looking into are more on the gap between protocols (Gunnar Andersson to be confirmed)
- Glenn: I will review the document and possibly add the standpoint of commercial fleet management
Data Categories
- Kevin I created a wiki page to gather information on data categories, link: https://at.projects.genivi.org/wiki/display/DIRO/Data+bundles
- this page is oriented towards the gathering of data categories (called "bundles") and the related use case
- /TODO/ all participants add your inputs on data categories in the data bundles wiki page
Monday, November 18
Participants
Glenn, Amir, Gunnar, Kevin, Keith, Benjamin, Guru, Steve, Philippe
Minutes
Debriefing of Cloud & Connected Services sessions at the tech summit
Project Charter
- Keith: revisiting the project charter would be helpful, in particular identifying the deliverables
- /TODO/ Philippe revise the project charter
Data Models
- Philippe: in my opinion, it turns out that we can proceed with VSS as the baseline data model
- Glenn: VSS is very helpful, there is a need to have a better awareness about having a data model, there is a need to present if in next year's conferences
- Glenn: we need to develop the data required by different use cases
- Gunnar: we would like to put most of the BMW data model into the VSS work
- Glenn: will be bring Ulf who is working with W3C in the conversation
- Philippe: could the identification of use cases be done using the ones identified in Kevin's presentation of the communication infrastructure ?
- Kevin: yes, use cases are also related to the data categories
Data categories
- Gunnar: we need to make categories of data w.r.t. sensitivity of data, privacy, different laws and juridiction
- Kevin: I propose to create a wiki page on data categories
- Benjamin: agreed it does make sense
- /TODO/ Kevin create a wiki page on data categories (for next week's call)
Data aggregation
- then Gunnar presents his notes from the workshop
- Gunnar: VSM we should look over this project , this is about combining the signals into other set of signals
- Glenn: data aggregation is one of the most important use case from our commercial fleet management business
- Gunnar: this relates to vehicle-side algorithms to aggregate data
Miscellaneous
- Gunnar: there is a need for exchanging (short) video in a commercial vehicle context
- Gunnar: it might be wortk looking at the use cases identifed in the 5G whitepaper (several whitepapers listed here, need to find relevant one(s))
- Gunnar: when it comes down to controlling a vehicle, specific remote APIs are likely better than data exchange
- Gunnar: there is a need to look at DDS
- Philippe: this is rather for the GPRO project backlog
- /TODO/ Gunnar & Kevin add notes from the tech summit sessions to the page of wiki minutes
System design
- Kevin: in this session, we discussed actually more fundamentals questions than the design itself like privacy, authorization, etc.
- Philippe: I would recommend we start using the term "policies" when we want to describe the constraints / attrributes attached to a category of data
Data models gap analysis
- Philippe: the participants need to review the google doc containing the gap analysis deliverables
- discussion on how to bring the doc to completenss
- /TODO/ Benjamin write an intro + a conclusion to the doc based on the list of bullet points
- /TODO/ Gunnar write the section on Vehile HAL, i.e. on how to connect the Vehicle HAL to the vehicle data capturing infrastructure
AOB
Tuesday, November 12 - Tech Summit - Towards a common data model
- Partial capturing of the Q&A
- slide deck
- Geotab - Glenn: gives the example of the lidar sensor that can spot available parking spots, can it be part of the VSS ?
- Benjamin: yes, we can model it as an extension
- Gerald: from the lidar, you get only a low-level information which is the distance-to-obstacle and if you look at the map, parking spots are not identified, much more processing is required, IMHO it is a very complex example
- Glenn: the purpose was to propose a use case outside the transportation itself
- Gunnar: reminds the audience about the dynamic agents concept that was pushed by JLR (capabilities to execute some highly-constrained code that can be uploaded on the vehicle)
- Keith: question on sensor description, how to push the VSS to an official standard like SAE/ISO or IEEE, how do we get broader adoption for VSS ?
- Glenn: we need to do dissemination among more OEMs
- FCA: we are looking into this, we are exploring it
- consultant: OEMs are afraid to lose their "secret" data, to get better value for the car, we should look at what other in telecoms did, in my opinion, there is a no secret data
- Gunnar: we need to understand how to write a contract for a data
- Keith: no data is secret, OEMs want to own the data and sell them
- X (consultant): telecom industry tried to keep the data proprietary but failed
- Keith: reminds the safety critical aspects of data
- Keith: why do google move to autonomous driving ? because of the data, Google wants to be in the revenue stream
- Y: VSS is a human readable description, OEMs are currently concerned by the telephone bills, is VSS the right modeling approah ?
- FCA: data means also taking commands from the cloud to the vehicle
- discussion on engineering units
- what are the data actually representing the driver's action on the accelerator pedal ? pedal position or pressure on the pedal ?
- Benjamin: should we include in VSS the AD(Autonomous Driving) vehicle sensors and the brand new IVI features ?
- Alex: I disagree with the way the VSS tree is currently described, I would prefer everything to be described in the vehicle data tree, no proprietrary extensions
- X (consultant): we need to push something useful and comprehensive enough to get accepted
Monday, November 5
Participants
Gunnar, Kevin, Keith, Gerald, Benjamin, Guru, Steve, Philippe
Minutes
Monday, October 28
Participants
Gunnar, Kevin, Keith, Gerald, Benjamin, Guru, Steve, Philippe
Minutes
Connected Services planning of tech summit
Session One: towards a common data model
- Introductory slides and more slides to guide a workshop:
- Define overall goal (* of workshop)
- find a common data model
- include questions to trigger discussion
- the goal of the project itself should be in the earlier overview presentation, but can be repeated
- /TODO/ Keith review the CCS project charter
- Outline from before:
- starting from the data in the vehicle
- VSS intro, AA Vehicle HAL intro
- Report on the gap analysis (Sensoris, etc.)
- Need for consolidating the models: Ask OEMs & participants about it
- Anticipate and think about answers to questions/concerns that come up: (Many new people, some might start from scratch with this standardization idea)
- Hard to standardize on CAN signals
- (find useful subset (already exists in diagnostics),
- => discuss the need for standards on the developer "API" level anyway.
- VSS can be used on CAN/network level but also on the higher levels.
- => Discuss the need to minimize (but maybe not eliminate) data translation across network -> local apps -> cloud → data/statistics users
- Someone might ask: Does the work cover remote Unlock car, start engine, and similar functions?
- =>There is definite resistance to this. Discuss reasons why? Is it security concerns?
- => Do open specifications have a (negative) relationship to cyber-security ? Prepare answer on this.
Keith: IMHO we need to explain we do not look at the downloading of data from the cloud to the vehicle, just upload of vehicle data, it might help breaking some resistance to the standardization of an upload link
- Concerns about that standards may be linked to OEMs being forced to give away data
- -> A controlled (standard) approach will allow handling this likely better than a "wild west"
- CCS can focus, if participants prefer, not on not only technical standards but also anticipate a framework for the commercial contract setups that control the actual live data that is exchanged at any time. In other words: Technical standards are not equal to the data exchange itself
- How to define data items is not the same as the actual data (sensor measurements, user data)
- How is access managed?
- => Fine-grained access control for applications and users of data
- Prepare presentation of access control approaches (technical means)
- => Access control in a Neutral-Server environment (probably input from Kevin)
- Before 10:45am-12:15pm (towards a common data model session) there is the project introduction session. Make sure to address some of the concerns there.
- Data on the internal vehicle level is not the same as outside the car. Both have potential for some standardization but can be done on only one of them too.
- How the choice of technical specification can affect the end result (performance, feasibility of ...)
- Policies for future-proofing (new versions of protocol specifications, and of the VSS itself).
- Discussion: What parts of the VSS-based ecosystem is based on a standard database of named signal, and how large parts are proprietary extensions?
- Example: Oil consumption unit depends on location.
- Units - VSS has losely specified an approach for this as follows:
- Prefer international (SI) units.
- But in some cases non-SI unit is the defacto standard in automotive.
- And in some cases there are no agreed standards.
- Question: Are there some signals that should allow for more than one unit to be fair to all OEMs ? E.g. km/h and miles/h. Some that have a preference extremely strong in some markets? Notable, drivers are in different countries so on the user HMI level, car manufacturers already do a lot of conversion to local preferences. So the problem would remain only on standardization on in-car networks
- /TODO/ Benjamin, prepare material with input from the above, and we can do a final review next week
- Present Gap Analysis results is also a part of this
Abstract of Session One (for reference)
The GENIVI Cloud and Connected Services project intends to define an end-to-end reference architecture between vehicle sensors and the back-end cloud. Essential to this architecture is agreeing to a common method of representing in-vehicle data. This session will provide overviews of the Vehicle Signal Specification (VSS), used in the definition of the W3C automotive working group. It will look at gaps existing with other data models and in the in-vehicle portion of the architecture, explore data in an Android Automotive context and consider ways of consolidating vehicle data modeling approaches
Session Two: The Value of Vehicle Data to Enterprises., Moderated by Keith (LGE)
- FIRST: Keith -- create short intro. (needs input on what the speakers will include, i.e. needs their slides)
- Connection to AUTOSAR? Keith is our liaison
- Session plan:
- Blueprint for Vehicle-Oriented, Data Strategy (Kevin Valdek, CTO, High Mobility) 30 min?
- Unlocking the Value in Vehicular Data Using Analytics (Amir Sayegh, Geotab) 30 min?
- Can we have a coffee break here? Yes (possibly local to this track) Coffee break - plan for 30 minutes
- Industry Standards for Advanced Vehicle Data (John Schmotzer, Emerging Analytics Connected Vehicle Manager, Ford) 30 min?
- Q&A: 30 minutes,
- Coffee break - plan for 30 minutes
- Total time : 150 minutes
= total 30 minutes remains for additional discussion time. - Should be adequate, if we can have the coffee break as planned _before_ the last talk.
Session Three: Capturing Vehicle Data and Communication Framework Design (Moderated by Kevin Valdek, High Mobility)
- Kevin is preparing his inputs following call with Gunnar and Philippe scheduled on Thursday last week
Gap analysis deliverable readiness check
- wiki page is in good shape, some additional inputs or reviews needed
- /TODO/ participants review Benjamin's text on CVIM
- Gerald reported that it seems HERE entered also the field of vehicle data and include also the neutral server concept.
- link: https://developer.here.com/blog/open-location-platform-2.8-release
- Access to vehicle sensor data - the Marketplace can now act as a secure, neutral, GDPR-compliant hub for data consumers to gain access to vehicle data from participating Automotive manufacturers
- Consent management - ensures privacy for owners and drivers of vehicles, by allowing them to grant and revoke consent for specific third-parties to access data generated by their vehicles
- ISO 20078 <https://www.iso.org/standard/66978.html> "ExVe" compliant interface - simplifies platform integration for data providers and consumer
- /TODO/ Guru review Here open location platform content and determine whether this is relevant for CCS and has an impact on our work, identify also a list of questions to be asked to the Sensoris project lead during the 2nd call we are considering have with him
Monday, October 21
Participants
- Stephen L
- Steve C
- Benjamin
- Don Dulchinos
- Gerald
- Gunnar
- Philippe
Minutes
Confirm Tech Summit agenda and participation
Current working schedule:
Tuesday 12 November
10:45am-12:15pm 16:45-18:15 CET | Workshop | 90 | Cloud & Connected Services Workshop Session One Toward a Common Vehicle Data Model (Moderated by Benjamin Klotz, Gunnar Andersson) Abstract: The GENIVI Cloud and Connected Services project intends to define an end-to-end reference architecture between vehicle sensors and the back-end cloud. Essential to this architecture is agreeing to a common method of representing in-vehicle data. This session will provide overviews of the W3C Vehicle Signal Specification (VSS), look at gaps existing with other data models and in the in-vehicle portion of the architecture, explore data in an Android Automotive context and consider ways of consolidating vehicle data modeling approaches. For reference but not posting to the public schedule: - starting from the data in the vehicle
- VSS intro, AA Vehicle HAL intro
- report on the gap analysis (Sensoris, etc.)
- need for consolidating the models: ask OEMs & participants about it
- moderated by someone from W3C (plan A) (Benjamin to look for flights, sync up)
Gunnar (plan B)
|
1:45pm-4:15pm 19:45-22:15 CET | Workshop | 150 | Cloud & Connected Services Workshop Session Two The Value of Vehicle Data to Enterprises (Moderated by TBC Need suggestions for moderator) Abstract: OEMs, suppliers and other industry participants have begun to realize the tremendous potential that in-vehicle data affords to their organizations, their customers and their partners. During this session, participants will hear from Ford, High Mobility and Geotab about the opportunities that vehicle data offers and how the industry can use data analytics to solve real challenges and offer new services. Presentations: - Blueprint for Vehicle-Oriented, Data Strategy (Kevin Valdek, CTO, High Mobility)
- Unlocking the Value in Vehicular Data Using Analytics (Amir Sayegh, Geotab)
- Industry Standards for Advanced Vehicle Data (John Schmotzer, Emerging Analytics Connected Vehicle Manager, Ford)
CONFIRMED all above
|
| Wednesday 13 November |
8:30am-10:00am 14:30-16:00 CET | Workshop | 90 | Cloud & Connected Services Workshop Session Three Capturing Vehicle Data and Communication Framework Design (Moderated by Kevin Valdek, High Mobility) Abstract: An essential part of an end-to-end, vehicle to back-end cloud reference architecture is the technologies used in the communication framework between vehicle and cloud. This session will focus on existing standards, solutions and gaps in that essential communication framework to roll-up into the planned reference architecture. For reference but not posting to the public schedule - Kevin as moderator is plan A – Kevin can do it remotely
- Following up with Kevin this week for details
|
Gap analysis document
Reference Design
- Feedback from all other participants
JIRA/sprint review
AOB
Stephen:
Interesting to note CAN XL standard https://www.can-cia.org/groups/technical-groups/technical-committee-tc/ig-layer-12/sig-can-xl/ being developed now.
Monday, October 14
Participants
- Benjamin
- Keith
- Don (OCF)
- Gunnar
- Philippe
- Steve
Apologies
Agenda
- GENIVI technical summit preparation
- Gap analysis deliverable - review feedback of draft sections on VSS, Sensoris, etc.
- Design concerns wiki page - more feedback
- Sprint & backlog review (end of Sprint 4)
- AOB
Minutes
- GENIVI technical summit preparation
- Philippe shows the content that will published on Wednesday this week at 12:00 US Central time
| Tuesday 12 November |
| 9:20am-9:40am | Plenary | 20' | Cloud & Connected Services Overview - Gerald (Bosch) will deliver the overview |
11:00am-12:30pm 17:00-18:30 CET | Workshop | 90' | Cloud & Connected Services - Data Model standardization - towards a common model - starting from the data in the vehicle
- VSS intro, AA Vehicle HAL intro
- report on the gap analysis (Sensoris, etc.)
- need for consolidating the models: we need to ask OEMs & participants whether the consolidation is mandatory or nice to have
- moderated by someone from W3C (plan A) (Benjamin can do it remotely), Gunnar with VSS stuff (plan B)
|
2:00pm-4:30pm 20:00-22:30 CET | Workshop | 150 | Cloud & Connected Services - Value of Data to Enterprises - company presentation by high mobility (Kevin) Confirmed (Kevin will attend remotely, timeslot blocked on his side), what would be the focus on the presentation ?
) - company presentation by Ford (John) on value of data To Be Confirmed topic: engaging the automotive supplier ecosystem and beyond (i.e., other industries) using vehicle data
- company presentation by geotab (Glenn)Steve contacted Glenn who is interested in making a presentation on this topic: aftermarket vehicle data use case
- followed by moderated open dialog with the automotive industry participants
- moderated by TBD: we need a volunteer, please contact Philippe
|
| Wednesday 13 November |
8:30am-10:00am 14:30-16:00 CET | Workshop | 90 | Cloud & Connected Services - Vehicle Data Capturing - Communication Framework Design - Kevin as moderator is plan A
- Kevin can do it remotely
|
- Philippe: we expect to have some kind of BoF meetings on Wednesday afternnoon
- Gap analysis deliverable - review feedback of draft sections on VSS, Sensoris, etc.
- Gunnar: has questions on Guru's section on Sensoris, will set up a dedicated teco for it
- Benjamin and Don reminded about their TODOs (see last week's minutes below)
- Big Picture & Design concerns wiki page
- Gunnar: presents the page to Gerald, Kevin and Benjamin who were not at the previous meeting
- AOB
- Keith would like to invite a colleague from LGE in Korea
- Philippe: no issue, please invite him
Monday, October 7
Participants
- Kevin
- Don (OCF)
- Gunnar
- Philippe
- Steve (partial)
Apologies
Agenda
- GENIVI technical summit - content planning, additional information
- https://grpc.io/: is this relevant for CCS ?
- SAREF presentation review
- Gap analysis deliverable - review feedback of draft sections on VSS, Sensoris, etc.
- Design concerns wiki page - more feedback
- Sprint & backlog review (end of Sprint 4)
- Enterprise data value: new topic to be added to the project scope ?
- AOB
Minutes
- GENIVI technical summit - content planning, additional information
- registration link is at https://www.eventleaf.com/TechSummit19ent
- content will be finalized this week, 3 tracks are planned: Connected Services, Android Automotive, Cybersecurity
- quick poll on who will attend
- https://grpc.io/: is this relevant for CCS ?
- g stands for Google
- Gunnar: we look at Google rpc in the GPRO project when we did the survey on communication protocols, look at the list of technologies we looked at and the GPRO white paper
- grprc is built on http2 which never took off, people are rather working on http3, gprc is likely well suited on web technologies, it is one of many choices that are out there
- from my perspective, it is relevant but not more relevant than anything else, when it comes to protocols that are not REST oriented protocols, MQTT has more relevance
- SAREF Automotive - https://www.w3.org/2019/09/trans-data-ws/SAREF.pdf
- Benjamin reviewed the paper presented at the W3C workshop on data models for transportation, his findings on SAREF automotive are here
- use cases identified in the paper are
- Platooning
- Automated Valet Parking (AVP)
- Cooperative Perception Service (CPS)
- Vulnerable Road Users (VRUs)
- Philippe: these use cases fit well into the the future, i.e. much beyond a 5-year horizon
- Philippe: IMHO currently in the CCS project, short term is represented by the current High Mobility offering on the market, and the on-going work is rather looking at the mid-term,
- Kevin: it would be good to ask OEMs what use cases are more tangible to implement
- Gap analysis deliverable - review feedback of draft sections on VSS, Sensoris, etc.
- Philippe: I reshuffled the pages a little bit and the draft deliverable appears now in the main CCS page table of content, look at Vehicle Data Models - Overview & Gap Analysis deliverable
- Philippe: Guru and Benjamin have provided their inputs, Guru added his inputs in the wiki, I added the link the CVIM doc prepared by Benjamin to the wiki page
- Kevin: I reviewed the draft document on VSS prepared by Gunnar, the intro is not specific to VSS but fairly general
- Gunnar: agreed, we need to ask again for the review work the VSS doc
- /TODO/ Daniel/Benjamin/Guru/etc please review the VSS doc
- Gunnar: I will review the Sensoris section written by Guru
- Kevin: I will review the CVIM doc prepared by Benjamin
- Gunnar: Steve pointed us to two other data exchange standards from the Open Group
- I found them to be mostly similar or related to what OCF might be doing. I also found them to be fairly thin specs in particular O-DM but the messaging standard might have something useful
- We are wondering what OCF think about it:
- Is it used by OCF or not ?
- Is it out of scope and covering other unrelated things ?
- or is it overlapping, but OCF have created other standards to cover this instead ? (It might of course boil down to a simple XML vs JSON preference...)
- Unknown User (don.dulchinos) Do you have any feeling on how well these standards are adopted?
- /TODO/ Don review these data exchange standards from the Open Group
- Gunnar: we should consider adding some info on SAREF and the Open Group work in the gap analysis deliverable
- Big Picture & Design concerns wiki page - more feedback
- link added to the CCS main wiki page
- Gunnar: the topics / issues in the design concerns page should be transformed soon in a set of work items described in jira, Kevin has provided his comments already, other comments are welcome until next week call
- I will also review the existing tickets in Jira that might be relevant to the topics described in the design concerns wiki page
- Philippe: I will projectize this in Jira starting next week
- Sprint & backlog review (end of Sprint 4)
- target date for producing the gap analysis deliverable is 8 November, we will present it at the tech summit
- /TODO/ Gunnar ping Michael Ger to get a Cloudera architect into CCS project
Daniel said he is "back" and should start working on it this week- /TODO/ Gerald please do
see above design concerns, projectization will be developed further starting next week
done, see above- no news from Sebastian, Gerald Spreitz can you help ?
- Enterprise data value: new topic to be added to the project scope ?
- Steve: when I met with Ford last week, one of their concerns was how to make value out of the vehicle data
- this is a gap in the CCS project charter according to Ford and GM perspective
- Steve: it seems to me that these organisations do not fuly understand yet what means a shared platform for capturing and distributing vehicle data
- Gunnar: one of the criticisms we had with the European OEMs concept of neutral server is that the OEMs will control all the vehicle data, we do not understand what the enterprise data value means for for the US OEMs yet,
- Steve: theUS OEMs seem to be looking for which strategic partnerships they need to build for the vehicle data
- Kevin: what kind of uses cases are foreseen , what kind of data rates are considered ?
- Philippe: I would suggest to ask high mobility (Kevin) to deliver a company presentation at the tech summit because HM business in my opinion exemplifies what the enterprise data value is about
Monday, September 30
Participants
- Ondrej (OCF Cloud Services)
- Kevin
- Guru
- Keith
- Gerald
- Don (OCF)
- Steve
- Gunnar
- Philippe
Agenda
- OCF (Open Connectivity Forum) Cloud Services WG introduction
- GENIVI technical summit - announcement
- Design concerns wiki page - feedback from participants
- Gap analysis deliverable - status of drafts
- Sprint & backlog review (skipped)
Minutes
- OCF (Open Connectivity Forum) Cloud Services WG introduction
- Ondrej works for Kitsler, a sensor manufacturing company https://www.kistler.com
- Ondrej shows a slide deck on OCF Cloud WG, Device API for Cloud Services, Cloud API for Cloud Services
- Philippe: IMHO the OCF cloud services are to be added as a possible cloud-based framework for poc development & end-to-end integration in the later stages of CCS project
- Gerald: can we possibly clarify about the mapping of the OCF cloud services to the automotive use cases ? with vendor A = BMW and vendor = GM for instance
- Ondrej : my car is one of the devices, when I sit in my car, I can see my local (home) data
- Guru: is there a limitation on the number of vehicles connected ?
- Ondrej: the implementation is fully scalable
- Ondrej will share the slide deck and the API specifications with GENIVI (how to share these materials with the CCS participants is being checked)
- GENIVI technical summit
- Philippe: announces the date and location for the tech summit - Tuesday 12 and Wednesday 13 November, in Troy, Michigan, USA and outlines the tech summit content that will focus on Android Automotive SIG, Cloud&Connected Services and Cybersecurity for the Vehicle
- Steve: has spent last couple of weeks reaching with Ford, GM and FCA, metwith Ford alrealy and will meet with GM on Friday this week and later with FCA
- Steve: the CCS project charter is close to what Ford thinks about where the market is going
- Steve: our objective is to engage those OEMs (and their suppliers) in working sessions about the CCS project
- Philippe: please all of you put a note in your calendar and check your travel plan ! thanks
- Design concerns
- Kevin provided his inputs as comments to the wiki page, look at Big Picture & Design concerns, comments are reviewed online
- Kevin: the wiki page is a very good thread opener
- discussion on Kevin's first comment on the protocols, actually rather on the different ways of performing the data capturing from the vehicle to the servers
Gunnar: on the protocol for Vehicle => 3rd Party Servers, we need to go out and survey who are the players in the industry and which models of data capturing are acceptable
Kevin: : it's a bit of a different case, it could be brought out separately from the in-vehicle infotainment system to third-party servers, because that's the route the data would take.
Gunnar: presumably, yes, it is like an app getting a direct connection and maybe we leave that out the scope and say that's not part of our infrastructure. But at the same time in order to cover all cases, we should consider these points, see what they mean for us and see if we can relate to them in some way in our specifications because otherwise companies tend to work around it anyway, and they end up doing ad hoc things that we might be able to avoid if we actually cover it in our discussions
Kevin: actually vehicle to third-party servers connection comes down to connecting to devices or devices in general like a smartphone, this is the obvious example of having a direct access from my mobile device or an application in a mobile device to the onboard systems, that's more of the typical MirrorLink type of solutions, but it is true there is a lack of data protocols there and if this is VSS that the Vehicle Gateway is using and if these are VSS data that are being uploaded to the cloud, then of course if there is a direct connection with Wi-Fi or Bluetooth from smartphone and the vehicle, it makes sense that the VSS model or a certain data model is used there as well.
Kevin: One more comment in the OEM server to neutral server part, we should add a connection from OEM server to third-party servers or developers as in general the OEMs will provide an API both for neutral servers who serve as intermediaries and directly to third parties, that's already starting to become a common practice and should be considered.
Kevin: I have another comment that relates to the question Gerald mentioned with big data earlier, in my opinion there are certain (or maybe even larger) design impacts of the system considering personalized data, which are more like small data chunks with high frequency and high velocity rather than big data type of sharing between two different services. Probably the system block-diagrams are to look quite different at some components level.
Kevin: in my second set of comments, I added some thoughts about the third-party expectations. This is nothing groundbreaking but more practical things. The callback feature is one thing that third parties expect. Third parties rely a lot on quite up-to-date data. That means that sometimes even one minute old data is not that good because of a business case where you need to act fast on certain data or because of the user experience for instance in electric car charging where you want to get an event as soon as charging is complete.
Kevin: with the VSS and the W3C, I've seen websocket protocols are considered, and the system design should have (when we have a Rest API or an SDK) a callback system with web hooks, this was mentioned also in the OCF presentation. Today a lot of these OEM systems that do provide data don't provide callback systems in general and it is a bit of a headache for third parties. Typically there are different request limits that can be hard to meet at the same time, if there are no webhooks, it creates a problem because third parties want to get the latest data at the same time and the OEMs might limit them with their request limits because otherwise they get too much traffic to the servers. This kind of problem needs to be solved with technology and should be part of the system design we are considering
Kevin: in the last two steps of my second set of comments, I added a few things about the fleets. In general with data sharing, there is a general consent from the owners of the car, this is more for personal vehicles, when it comes to fleets, there are different considerations. If a fleet owner has 1,000 vehicles for instance, the third party application and the fleet owner are the same entity, in that case how that is handled and how the data querying is handled probably should be also part of the overall system design.
Gunnar: I agree with your views, IMHO we should have a list of typical applications / use cases, i.e. something to drive where we need to do different solutions. As you said, the solutions might be different for fleet management applications and user oriented applications. Also what you said about the Big Data statistics being a different one from user-centric data.
Kevin: agreed, I could add a list of use cases in the wiki page trying to cover some different verticals.
Philippe: we need to ask again Cloudera about their inputs of the system design for gathering vehicle big data
/TODO/ Gunnar set up a call with Cloudera to get their inputs of the system design for gathering vehicle big data
- Gap analysis
- Gunnar: shares the draft sections on VSS available at this document and asks participants to review it
- Guru: wonders whether building a huge wiki page containing all the sections of the gap analysis is the right approach
- Gunnar: for the time being, in my opinion, the sections for each project should be structured according to the wiki page table of content
- Guru: will finalize the draft sections on Sensoris accordingly in the document he initiated and will call for a review in one day or so
Monday, September 23
Participants
- Gerald
- Stephen
- Don
- Steve
- Gunnar
- Philippe
apologies: Guru, Kevin, Ondrej (OCF)
Agenda
Minutes
Monday, September 16
Participants
- Keith (LGE)
- Kevin
- Guru
- Don
- Gunnar
- Philippe
apologies: Gerald
Agenda
- Roundtable
- W3C workshop on data models for transportation - Debrief (Steve) (skipped because Steve could not attend the weekly call)
- Sync opportunity with Autosar (Keith, Steve)
- Gap analysis deliverable
- AOB
- system design
- OCF cloud WG
Minutes
- roundtable
- Keith: LGE has an automotive component division, not quite like Hitachi or Yamaha though
- Keith is the only person working on Autosar at LGE, US, Keith's role is to be fully active in Autosar
- 2-3 months ago Keith started raising the awareness around Autosar in the Silicon Valley
- Sync opportunity with Autosar
- during this week's meeting with Steve Crumb, the slides he showed sounded very familiar to me because they were close to what I presented the Cloud Services WG to Autosar sometime ago
- the logical next step is to come with a single standard, in my opinion there is potentially a lot of synergy on the topic of cloud services between the two consortia
- LGE has a particular interest in cloud services
- the possible working mode Steve and I identified would be that Autosar endorse whatever will come out from the GENIVI work
- Gap analysis deliverable
- Kevin will make his contribution by Thursday
- Guru shows the (Word) document he is working on for the Sensoris part, he will import it into Confluence soon
- Philippe recommends Guru to maintain a list of questions to Sensoris, we will have an opportunity to pass them over to Sensoris project lead during the second call scheduled in almost 2-week time
- Philippe changed the status of the Jira tickets related to the gap analysis work
- Gunnar shares the content outline for the VSS part of the document
- Don asks about the status of VSS work and would like to sync with Ted Guild on W3C work
- Gunnar suggests to set up a call with Ted
- System Design
- Gunnar introduces the proof-of-concept for performing the translation from VSS to Franca to Some/IP, bottom-up works better for me
- Philippe: please look at the minutes of the AASIG Vehicle Data API project
- Gunnar; the question is whether we make direct conversions from data specs (like VSS) to protocol like REST or do we go through Franca ?
- Philippe: in my opinion, we should have a list of open questions
- Gunnar Andersson: /TODO/ Gunnar initiate a list of open questions related to system design
- Kevin: I can add to the initial list questions coming from our experience with 3rd parties at High Mobility
- Adjourned: 18:00 CET
Monday, September 9
Participants
- Guru
- Don
- Benjamin Klotz
- Gunnar
- Philippe
Agenda
- Gap analysis deliverable status
- Sprint & backlog review
- Closing of Sprint 3
- additional work items related to Analysis epic to be added to Sprint 4
Minutes
- Gap analysis deliverable status
- assignment of sections to contributors: Benjamin, Guru, Gunnar, Kevin
- Jira tickets updated accordingly CCS-54 to CCS-58
- Sprint & backlog review
- Benjamin: this relates for instance work in W3C Data TF, look at https://www.w3.org/community/autowebplatform/wiki/Data_tf
- Gunnar: this is not gap analysis, this is something different, it relates to the glossary / definition of terms, we should create a wiki page describing what these terms mean
- the W3C wiki page above is one input for this ticket
- assigned to Gerald for adding a description of the work item
- discussion on the functional architecture reference diagram
- Gunnar: the picture is not final yet, it is just an example
- Philippe: we should ask Cloudera for their vision of the big picture for the big data gathering
- created and assigned to Michael Ger
- Sprint 3 completed and Sprint 4 started
- OCF update
- Don gives an update of his discussions with OCF leaders
- there is a discussion with OCF about the level of membership that should be propose to vehicle OEMS in order to recruit them
- Don would like to have the OCF Cloud WG lead attend next week's call, an agenda item will be added for this
- Philippe: provides clarification on the expected use case to be shown at CES2020 from GENIVI standpoint
- last agenda item: Gunnar Andersson - can you add your notes, since I had to switch over to another call ? thanks
Monday, September 2
Participants
- Gerald Spreitz
- Benjamin Klotz
- Gunnar
- Philippe
apologies:
Minutes
- Call with Sensoris
- very good and friendly atmosphere in the call
- Prokop seems to be quite open for a coordination discussion
- He gave a rough overview od Sensoris including business cases and system architecture, business cases are OEM-related and do not intend to provide an open access to the vehicle
- Sensoris objective is to describe what the car is sensing for the world surrounding it, e.g. number of lanes, road works, traffic signals, outside temperatue (as part of a data set called climate, temperature and weather)
- Gunnar: VSS (was described by Prokop as...) focusing more on the in-vehicle data and SensorIS more on modeling the outside world that the car drives through. We have not been through to compare the data models during the call but there seems to be is some overlap in some areas but not in other like map-related data and data with a location-based focus
- Gunnar: License - Sensoris do not permit derivatives (Using CC-BY-ND). It could be interpreted as they are only body that can launch Sensoris specs, but also means that flexible reuse is difficult. They propose openess to anyone who wants to change the SensorIS spec.
- Unknown: "This is the way they work with Japanese projects"
- Gunnar: Sensoris welcome any proposal that would make their data model more complete, but this proposal would become part of the Sensoris data model because of their licensing approach
- Prokop shows an architecture picture (that includes GENIVI and AGL as a kind of "technology choices" inside of vehicle. Explanation was made about GENIVI Alliance actual scope and current projects.)
- Sensoris main purpose is to make sure that OEMs are not delivering different data, they should deliver the same data
- Prokop said it is out of scope of Sensoris to describe the system and software architecture, however they care about the "how" (e.g. protocols) to some extent
- Gerald: motivation for having a common data model in Sensoris is for instance the wipers model (e.g. 3 bits on the CAN bus)
- Gunnar: Gerald had a quite long discussion with Prokop about how OEMs provide the data, Sensoris does not care about the software architecture OEMs will use, more that the data that comes out is according to some shared standard (so that 3rd party developers can engage with more than one OEM).
- Gunnar: it seems they would reuse whatever architecture is defined by GENIVI
- Philippe: IMHO we should not fight for having an aligment of the data models, we could likely live with some duplicated representation for the data that overlap between Sensoris and VSS
- Benjamin: agreed
- discussion follows on GENVIVI scope vs Sensoris scope, VSS could be considered upstream w.r.t. Sensoris for the data defined in both projects
- another meeting will be scheduled in 4-week time, we need to do our gap analysis between Sensoris and VSS first
- will be added to the new sprint
- Scrum review
done
done
kept in the backlog, can be activated whenever Don provides us with an update on the OCF Automotive group work- done,
created, we might be interested in knowing more about the OCF cloud services framework in the future when we start looking at the available technologies for the framework
done
done, Gerald provided an updated version of the CCS block-diagram highlighting GENIVI components rather than Autosar framework
done- added to the sprint, new Jira ticket created
with blocking link added 
updated- Gerald: IMHO ISO ExVe is not too exciting, performing a gap analysis between ExVe is to be further discussed with Kevin,
- ticket is reassigned to Kevin to determine whether it is worth GENIVI buys the access to the specs from ANSI/ISO,
- Gunnar: then we should likely check the architecture proposed by ISO ExVe can be filled by some components we have identified in the GENIVI communication framework
- Review of remaining items related to Analysis epic in the backlog
actually done with the various surveys completed- tickets to be added to the next Sprint starting on 10 September:
- inputs for this work item will be given by the various gap analysis we have identified
- Gap analysis deliverable
- updated with subtasks for Kevin, Gunnar, Gerald, Guru, Benjamin, Daniel for describing their contribution to the deliverable, please refer to the GAP analysis outline draft
Monday, August 26
Participants
- Gerald Spreitz
- Kevin Valdek
- Don Dulchinos
- Benjamin Klotz
- Gunnar
- Philippe
Minutes
- Sensoris
- Gerald has contacted the project chairman
- timeslot proposed on Thu 29/8 at 8:30am for a sync call (expected participants: Gerald (call moderator), Gunnar, Philippe, Benjamin),
updated and closed - Philippe will send an invite once the timeslot is confirmed by Sensoris chairman has confirmed the timeslot
- discussion on the draft content of the gap analysis deliverable prepared by Kevin: Vehicle Data Models - Overview & Gap Analysis deliverable
- discussion on whether to include CVIM in the gap analysis, even though the results of Automat have not been exploited by OEMs which participated to the Automat project
- consensus seems to be there are some good stuff in CVIM that is worth appearing in the gap analysis
- Kevin, Gerald, Gunnar will start populating the deliverable in the wiki
- report from Don on OCF automotive
- Don was not able to sync with OCF staff as expected
- Don showed a slide deck about working mode with GENIVI
- next week agenda: backlog and sprint review
- misc. Benjamin's slide deck on his findings on CVIM are available there
- Adjourned: 17:05
Monday, August 19
Participants
- Gerald Spreitz
- Kevin Valdek
- Michael Ger
- Steve Crumb
- Don Dulchinos
- Guru
- Stephen
- Benjamin Klotz
- Gunnar
Minutes
- Kevin Valdek opened the meeting and welcomed participants.
- Discussing the organization of the gap analysis document...
- Kevin has produced an outline proposal in CCS-48.
- Gunnar has put some ideas into a wiki page
- How do we slice the document. One chapter per compared (data model) covering all characteristics, or one chapter per characteristics/topic covering all data models.
- some discussion, no obvious conclusion - but let's write an outline for clearer proposal we can decide on.
- SC: I would like to see a motivation chapter (for the GENIVI deliverable itself) to clarify document purpose and organization to the reader.
- Gunnar: Seems like we need to rework this a bit with the given input.
- Kevin: I volunteer to propose a (new, detailed) outline of the document
- Kevin will rework the Wiki page that Gunnar started on (because it is now kind of obsolete).
- Gerald: I can report on my action (CCS-28). Yes, overall we think CCC digital key is not much in scope. About car data, they are apparently not planning to create a data model / data taxonomy.
- Gunnar: OK so their main deliverable is requirements, or a design and a programming API? If a programming API, there can be some similarity between an API and a data exchange model, so I would at least take some look (for overlap) if it is an API.
- Gerald: ... yes it seems mostly requirements, use-cases, ... maybe more. I can check a bit more.
- Kevin/Gunnar: We should include them (any organization that falls a bit out of scope) in the report nonetheless, just to document that we have analyzed them and drawn a conclusion.
- Gunnar: Let's go through the JIRA tickets so we don't miss anything
- Kevin: I can do that, I will display...
- VCS 25 - Gerald: I sent a mail to Sensoris chairman about a joint meeting. No answer, need to retry this.
- CCS-50 - (W3C workshop) Steve Crumb: Yes it is confirmed that GENIVI will present the project on the W3C workshop
- CCS-2 This is Philippe's ticket and looks like a kind of overview tracking ticket (An Epic, sort of)
- CCS-33 - (SENSORIS vs VSS) Still open only because the presentation has not been published in the Wiki (Benjamin was reminded)
- CCS-25, Guru working on it. Keep it open.
- CCS-34 - Sensoris vs CVIM. We think we can close the ticket. Overlaps other analysis.
- CCS-38 - Steve Crumb: I have an update. I contacted ANSI via helpdesk. They require us a site license ($1200) to get the specification and we must put limits on the individuals that may read it. How many of you can get your own access via your organization?
- Kevin: I can read the ExVe document.
- Gerald and Guru: Also we can do that.
- ...BMW? (Benjamin can't go on audio, but we assume it ought to be accessible – follow up)
- SC: GENIVI might purchase it only for staff since the limited circulation requirement will in that case not allow us to distribute to the members. Steve will investigate this.
- CCS-39. Kuksa. Gerald will contact Sebastian to see status
- CCS-42. Automat Security analysis. Gunnar: I will try to make sure it is presented to the security meeting this Thursday, as requested. We assume this ticket stays open until then.
- CCS-47. Don: This is on track. I am talking to OCF tomorrow.
- Kevin: Any other business?
- Meeting adjourned
Monday, August 12
Participants
- Kevin Valdek (High Mobility)
- Gerald Spreitz (Bosch)
- Don Dulchinos (eonti)
- Gunnar (GENIVI)
- Philippe Robin (GENIVI)
apologies:
Planned agenda
- Gap analysis deliverable: document owner and contributors, table of content update
- Sprint 2 closing
- Various topics according to JIRA (SCRUM sprint review)
Minutes
- Philippe welcomes Gerald and Gunnar who are back from vacation
- Sebastien Schidlt sent apologies, he will try to join in a few weeks
- Gap analysis deliverable
- contributors
- Gerald: yes, Bosch will contribute, but gap analysis topic is TBD, I need to check with Guru
- Kevin: yes, HM will contribute, quite open about the topic(s)
- Gunnar: yes, I will contribute
- Gunnar: when performing the gap analysis, we will identify new topics to investigate in addition to filling the content of the gap analysis
- Philippe: we are in an agile process, no worries
- Philippe: we need to get Daniel and Benjamin in the loop
- Philippe: I added Androif Auto P.Car to the list of projects we will review, the analysis of P.Car is starting in the Android Automotive SIG Vehicle Data APIs subproject, from this work, we expect to get a gap analysis between what AA provides and what the OEMs need w.r.t. the vehicle data APIs
- /TODO/ Kevin, Gerald, Gunnar initiate a thread offline to decide on the topics assignment, created and assigned
- Philippe: IMHO we need to start writing now
- OCF update
- Philippe: welcomes Don Dulchinos representing OCF Automotive
- Don: introduces himself, Don is in consulting business - contracted to a security company which is interested in automotiven, tried the OCF Automotive project a year ago, set up an official liaison with GENIVI
- this year has done some work in April/May, OCF has started a cloud interface WG and wanted to set up a liaison with W3C and this took a long time
- in June there was an event in Colorado with JLR and W3C where they discuss to have a poc done, OCF has some interest in it but not strong enough
- then Don organized a call with with Ted and chair of OCF strategy,
- OCF got recently renewed interest in this automotive group: Honeywell (smart thermostats), interested in the connection to the home
- use case: someone is driving to his home and wants to get access to the home for setting thermostats
- Gerald: is there the reverse direction use case ?
- Don: not an Honeywell use case
- OCF is interesting in having a demo at CES 2020, there is also an interop event scheduled before in December
- discussion on demos
- Philippe: timeline for a Cloud & Connected Services project demo is for CES2021, we intend to work before on a gap analysis between various projects working on the vehicle data access and then we will scope and specify possibly a common framework for the vehicle data access and define APIs
- Gunnar: back to OCF demo, what are the companies investigating the cloud APIs in OCF ?
- Don: Samsung, Honeywell for instance
- Gunnar: IMHO OCF would consider joining the GENIVI showcase ar CES 2019, you would need to contact the organiser of this event
- Don: I will talk to the governance of OCF (i.e. companies like Samsung, Intel, Honeywell active in the cloud interface)
- Don: is GENIVI active with the W3C ?
- Gunnar: yes, we are very much active, we have some kind of shared memberships, we work closely with them
- Scrum review
- VCS tracker https://at.projects.genivi.org/jira/browse/VCS
- Sprint 2 completed, Sprint 3 started
Survey CCC specifications and check for relevance- Philippe: IMHO the Digital Key project is not relevant for our CCS project, my interest is in the Car Data project; is it an active project within CCC ? are there deliverables available ? if yes is it worth looking at them ? if yes, we will determine later how GENIVI can access them. assigned to Gerald
- Assign gap analysis deliverable sections to ownerscreated and assigned (please look at the list of watchers who are expected to engage in the thread)
Survey SENSORIS specifications- Philippe: Guru would like to dig further into Sensoris and check on histograms in Sensoris (as a follow up of last week's discussion on automat )
- reassigned to Philippe for tracking ISO feedback on the access to the specs
- Sebastian was poked about providing an update on kuksa project
- Philippe: following last week's discussion on Automat, we need to point the GENIVI security team on Petar's overview, Petar added to the thread on assignment of the gap analysis deliverable sections
- assigned to Don, Don will provide an update on OCF on 26 August
review CCS project charter Done- contact Sensoris and ask for a sync call, assigned to Gerald
- Philippe: Gunnar and I need to go through the backlog and identify which other tickets are to be scheduled in Sprint 3
- Next week call
- Kevin will moderate it (thanks)
- main agenda topic is the assigment of gap analysis deliverables sections to contributors
Monday, August 5
Participants
- Gururaja N (Bosch)
- Kevin Valdek (High Mobility)
- Michael Ger (Cloudera)
- Philippe Robin (GENIVI)
Planned agenda
- 1- Update on automat project
- 2- Gap analysis deliverable: document owner and contributors, table of content update
- 3- Various topics according to JIRA (SCRUM sprint review)
look at active sprint & backlog of VCS tracker https://at.projects.genivi.org/jira/browse/VCS - 4- AOB
Minutes
Kevin moderates the call. Thanks.
1- Update on automat project
- Philippe reports the outcome of the two calls he had with Automat project participants last week
- Renault participant on Wed 31 July
- Automat cybersecurity WP lead (from Trialog) on Tue 30 July (A,ntonio)
- call wirh Renault
VW had the biggest budget share in the automat project (up to 25%)
the VW team involved was from the After Sales division (not from the R&D) and had a very short term vision of their needs
VW interest was on what happened in the vehicle before a failure using histograms, because they wanted to minimize the telecom costs for data transmission and therefore the quantity of data transmitted
there were actually two competing approaches for the vehicle data telemetry in the automat project: histograms for VW vs time series for Renault and Fiat (CRF Fiat R&D), the time series approach consisted in transmitting as many raw data as possible and processing them offline with big computers
At the beginning of the project, HERE proposed to use the Sensoris approach for gathering vehicle data but there was nothing like histograms in Sensoris and this approach was not investigated further
Renault did a proof-of-concept using a dongle to gather the data, VW and CRF decided to use existing ECUs (CRF used an existing TCU (communication box)
At the end of the project the OEMs decided not to continue.
- call with Trialog
Trialog pointed to the following position paper from BDVA (Big Data Value Association), Antonio was one of the main contributors to this paper based on his work in the automat project with a particular focus on the privacy for big data
Antonio mentions also the SAREF work on data ontologies pushed by EU whose motivation is to define and promote a European standard
Antonio mentions W3C / Web of Things (WoT) work whose charter starts with a statement on the ecosystem fragmentation
data privacy / security in automat
the interesting feature of the automat cybersecurity work is that it relies on existing standards (at the time of the project execution timeline)
Antonio said it would be good to review and restructure and update the automat cybersecurity framework according to recent US NIST work that takes into account the product lifecycle (to be shared witl the GENIVI security team)
- Conclusion on automat
Philippe's personal conclusion is that we have looked enough into Automat, unless we have an indication that Autosar will eventually use Automat CVIM as an input to their Cloud Services WG work, we in GENIVI should focus on Sensoris and VSS and ISO ExVe, we will also coordinate with Autosar on this
call participants agree
- Guru: said he will look further into requirements for histograms as defined by automat and check what was not supported in Sensoris
However the work done in automat on cybersecurity is good and relates to a big data architecture, Philippe recommends the GENIVI security team to review it and amend it w.r.t. US NIST work and extract possibly use cases to benchmark the MoRa tool
- Additional information (for Kevin and others)
- yet another proposal from VW
2- Gap analysis deliverable: document owner and contributors, table of content update
- Philippe: said he added P.Car (from AOSP) to the list of projects on Vehicle Data APIs we have to consider for the gap analysis
- P.Car is investigated in the AASIG Vehicle Data APIs subproject which started last week, look here
- Kevin: asked whether we will have one or several deliverables on gap analysis
- Philippe: said he is fine with both approaches, we need an executive summary to provide an overview of the results anyway, the format of the deliverables could possibly followed the one of the technical papers published earlier like GPRO white paper
- Philippe: we need to assign sections / deliverables to contributors and nominate a document owner, we will create the relevant Jira iickets in the new sprint that will start next Monday
- Kevin: said he (and his team) can do a contribution
- Guru: will check with Gerald whether Bosch can do a contribution
3- Scrum review
- VCS tracker https://at.projects.genivi.org/jira/browse/VCS
- commented and updated, subtask created on how GENIVI could access ISO specs created and assigned to Steve
closing remarks on automat project added
table of content for gap analysis doc reviewed and agreed, closing remark added (see above), done- we need to find an owner for the review of CCC specs
- we will close the sprint on next Monday 12 August
Monday, July 29
Participants
- Gururaja N (Bosch)
- Kevin Valdek (High Mobility)
- Michael Ger (Cloudera)
- Stephen Lawrence (Renesas)
- Steve Crumb (GENIVI)
Apologies
- Petar Vorotnikov (Visteon)
- Daniel Wilms (BMW)
- Gunnar Andersson (GENIVI)
- (Magnus, Melco - vacation)
- Gerald Spreitz (Bosch) (vacation)
- Philippe Robin
Planned agenda
- Review of the gap analysis deliverable content proposal (all)
- Presentation of a summary of the Extended Vehicle ISO 20078 standard contents
- Various topics according to JIRA (SCRUM sprint review)
- look at active sprint & backlog of VCS tracker
- JIRA kanban is here (Note, Kanban requires login. This browse project seems to currently work for viewing without login. Please request login if you are active participant....
Minutes
Kevin moderates the call. Thanks.
Gap analysis deliverable proposal
- Kevin: the proposal from Philippe is good and we already have most of the necessary information about each project to start
- Guru input:
- For the scope overlap/gap part we can bring in a table structure that we use for comparison
- We can define data characters from one project to another
- We can add information about how each project is being adopted by the industry
- Steve: when talking about deliverables, can we draw experience from the work being done in the Android Automotive SIG? Data topics are being discussed there as well.
- Philippe (added offline) : IMHO we could add P.Car as one of the inputs we need to review, to be discussed at the next meeting
Extended Vehicle ISO 20078 summary
- Kevin presents - key points:
- The ISO standard was published end of Q1 this year. It covers concept and specific requirements for OEMs to follow when providing car data access to third parties via web services
- The standard consists of three different parts:
- Content (resources)
- Security framework
- Access (consent flow)
- The standard does not enforce or recommend any specific car data protocols
- Should be useful for the CS project. To consider: if a deliverable should be compatible with the standard
- Documents are not available in the public domain, but licensed
- Guru:
- There’s a lot of info in the document that is useful: Authorisation, REST API etc.
- As the documents are not in the public domain. Does GENIVI have any type of collaboration with the ISO body, to get access to the information? How could it be done?
- Steve: will investigate if we can get a license
Monday, July 22
Participants
- Benjamin Klotz (BMW)
- Daniel Wilms (BMW)
- Gururaja N (Bosch)
- Kevin Valdek (High Mobility)
- Petar Vorotnikov (Visteon)
- Gunnar Andersson (GENIVI)
Apologies
- Michael Ger (Cloudera)
- (Magnus, Melco - vacation)
- Gerald Spreitz (Bosch) (vacation)
Planned agenda
- Update on CVIM / VSS data comparisons (Benjamin)
- Give feedback and approve project charter (all participants)
- Various topics according to JIRA (SCRUM sprint review)
- look at active sprint & backlog of VCS tracker
- JIRA kanban is here (Note, Kanban requires login. This browse project seems to currently work for viewing without login. Please request login if you are active participant....
Minutes
Gap analysis CVIM vs VSS
Benjamin: there is a lot of overlap between cvim & vss, it would be good to have someone from the automat project in the call
/TODO/ Philippe contact automat project manager & renault (since Renault is one of the project participants)
created
Sprint review
Monday, July 15
Participants
- Gerald Spreitz (Bosch)
- Michael Ger (Cloudera)
- Benjamin Klotz (BMW)
- Daniel Wilms (BMW)
- Ilios Galil (Obigo)
- Gururaja N (Bosch)
- Kevin Valdek (High Mobility)
- Petar Vorotnikov (Visteon)
- Gunnar Andersson (GENIVI)
Apologies
- Philippe Robin (GENIVI) (vacation)
- (Magnus, Melco - vacation)
- ...
Planned agenda
- review of OADF workshop minutes
- discussion on gap analysis deliverable content
- feedback on project charter
- AOB
- JIRA kanban is here (Note, Kanban requires login. This browse project seems to currently work for viewing without login. Please request login if you are active participant....
Minutes
- Presentation of Michael Ger and other participants.
- Gunnar: I'm hoping Cloudera will provide great input to the "Big Data" flavor of the technical architecture (which likely complements the on-demand protocols like W3C VSI).
- Gerald went through the minutes of the OADF meeting and through that described the different topics that are discussed in that forum.
- Guru presented analysis of current Sensoris work (Analysis document *)
- Benjamin presented AutoMat overview (slide deck *)
* please find links to material from the corresponding ticket in the JIRA project if not provided here.
Over time the material ought also be summarized as results of the project on the parent Wiki page.
Discussion:
- Michael Ger asked for login. Gunnar answered via email.
- Gunnar asked every presenter to update their corresponding JIRA ticket. If anyone is still having issues with login, email Gunnar and/or Philippe (on vacation this week).
- Benjamin's presentation was a little short on time. Decided to follow up next week on more detailed discussion on comparing the data models.
- Gunnar will create a ticket for "gap analysis" accordingly.
Monday, July 8
Participants:
- Benjamin (BMW)
- Gerald (Bosch)
- Guru (Bosch)
- Daniel (BMW)
- Petar (Visteon)
- Magnus (Melco)
- Ilios (Obigo)
- Gunnar
- Philippe
apologies:
Agenda:
Minutes
Monday, July 1
Participants:
- Benjamin (BMW)
- Gerald (Bosch)
- Guru (Bosch)
- Daniel (BMW)
- Kevin (High_Mobility)
- Gunnar
- Philippe
apologies: Petar (Visteon)
Agenda:
- Connected Services project charter
- ISO Extended Vehicle update
- W3C status presentation (Gunnar)
- CVIM review finding (Kevin)
- sprint review
Minutes
- Connected Services project charter
- ISO Extended Vehicle update
- W3C work status
- Gunnar presents the content of the following wiki page Connected Services: W3C VISS/VSI specification status
- discussion on proff-of-concept implementation of the VISS/VSI
- Kevin: is there a general plan/timeline for VSS Gen2 ?
- Benjamin: things will be delayed until the data models for transportation workshop has happened (scheduled on 12-13 September 2019)(look here)
- current WG charter started on May 2018 and will end on June 2020,
commented
- Sensoris
- Gerald shows a slide deck (link TO BE ADDED) from Bosch describing the status of Sensoris work
- Version 1.0.0 of the Sensoris specification will be released in the next days under CC-ND license
- data provided by Sensoris are not only about the vehicles but about the environment
- Version 1.1.0 will introduce the concept of "jobs"
- jobs = login jobs running for a certain period of time for a certain fleet of vehicles
- Gunnar: it would be good to identify how much overlap we have, shows the interest of different domain taxonomies
- Gerald: it would be good also to set up a call with the sensoris project once we have reviewed the Sensoris specification that will be published soon
- AOB
- telco schedule
- DECISION we will start scheduling the weekly call on Monday afternoons at 4pm CET next week (8 July)
- (added offline) cloudera and autonomic.ai which are based on US Pacific coast have been invited to join
Monday 24 June weekly call
Participants:
- Petar (Visteon)
- Gerald (Bosch)
- Daniel (BMW)
- Gunnar
- Philippe
Minutes
Monday 17 June weekly call
Participants:
- Petar (Visteon)
- Kevin (High Mobility)
- Benjamin (BMW)
- Gerald (Bosch)
- Ilios (Obigo) new participant
- Gunnar
- Philippe
Minutes
news
project charter
Philippe is preparing a project charter for advertizing our work, we foresee to breakdown the project into 2 stages
Stage 1 focusing on data, i.e. what we are initiating now with an objective to decrease the fragmentation of the ecosystem with several different approaches to describe the vehicle data and the communication framework to access them
Stage 2 focusing on compute, initial though is to look into cloud computing platform to support mobility and vehicle autonomy
- this charter will be shared with this project participants
- project tracker
- new participant
- Ilios introduced himself, he works for Obigo (Korea) and is their representative in Europe, he is based in Portugal-based representative attended the weekly call, he will be an observer for the time being, we had a discussion on the GENIVI Korea REG
- Gerald: asked Iliol whether there are public funded projects related to vehicle data in Korea
- project status
- we use the Jira tickets of the active sprint
- Kevin (High-Mobility) has reviewed the automat project Commoin Vehicle Information Model and presented his findings
CCS-26 - Survey AUTOMAT specifications (Common Vehicle Information Model - CVIM) In Progressthe model describes big data for connected cars, it is quite different from VSS, it goes very deep in data harmonization and proposes several layers like signals + measurement layer + data package layer
Gunnar: how is it compared to W3C
Kevin: I have not seen the same information than in W3C
Benjamin: we have a metadata Task Force in W3C, it is part of what next generation will do, we will have different kind of meta data, I would recommend to use the best practice, no competition now from W3C because there is no spec on this topic yet
- Kevin will summarize his findings in the wiki
- Petar (Visteon) is reviewing the automat project security model
CCS-32 - Read AUTOMAT security document In Progress - Gerald (Bosch) will spot the colleague within Bosch who attended the OADF conference, objective is to learn about the Sensoris specifications status CCS-25 - Survey SENSORIS specifications To Do
- Gerald: we need to make a comparison between the various data and protocol specifications
sensoris va automat ==> Gerald / Kevin (due date: 3-week time
vss vs sensoris ==> not assigned yet
vss vs automat ==> Benjamin (after submitting his thesis at the end of the month) , Gunnar
- Jira tickets created in VCS
Wednesday 12 June
Webinar on W3C Auto WG & GENIVI collaboration, VSS and Vehicle Data definition future steps
Wednesday, June 12, 1300 CEST
recording
Monday 10 June - informal call
(This call was not intended/schedule and therefore had no official invite. Therefore only had 3 participants, but we still made progress in discussing the needed steps and plans which will benefit the group for next time)
Participants: Kevin, Petar, Gunnar
Using a Etherpad at Mozilla we captured:
Security guidelines
1. Collect and write down Security guidelines
Idea to produce general security guidelines for different parts of the data-oriented parchitecture.
E.g.
- For this API type you should apply this type of security...
- For this data link ....
- Web services and Mobile App requirements are different
- High Mobility has input to give here
- GENIVI Security group could review, if we have something written down
(We discussed to prepare a draft result first since the Sec Team does not benefit from an undefined "open question" at this point)
Data categories
2. Analyze What type of data is needed for different data users?
- Recognize that fine-grained filtering is needed, and mechanisms for it
but here we rather mean:
- Define names of groups of data, and describe them.
...Service classes that need different refresh rates
- Define groups and needs
Data contracts
- Quick intro to discussion in W3C before. It came up because it relates to the definition of the needs of each party. These needs eventually get formalized into contracts and that is the link.
- Grab-bag of tools/questions/terms to define business contracts or more technical contracts
- A kind of checklist...
- Purpose: Simplify business contracts on data, and make sure details are not lost
Tech constraints, technology choices
What are the Technical constraints that are imposed by the used technology?
- Communication paradigm and protocols (pub/sub, req/res, REST, MQTT)
- Middleware and components standing in the data path - caching layer, service layer, service bus
Surveying existing initiatives
- SENSORIS -> await this week's meeting and announcements, and see the outcome before defining actions
- Automat - Data formats and types and similar things are in their specs. Read and consider alignment.
https://automat-project.eu/content/open-cvim-specification
- W3C protocol content (being worked on with participation in group - but summarizing and presenting to the group is useful). Also Webinar tomorrow.
- CCC - have we confirmed there is nothing to reuse or relate to or is this still active? Reconfirm with PR and GS.
- ... follow up with the rest of the list
Question: SENSORIS - which companies can lead the investigation and alignment?
Answer:
- not Visteon - Petar did not reach anyone working with it
- not HM -
- Bosch? ... figure out this list
Terms/names/definitions
Steps:
- Review existing specs, do they include a Glossary
- ... and/or just start writing it down
Collecting Information in Wiki to complete the picture
- Action for all: Populate the Wiki with useful links
- Example of things to link to:
Other Projects
Specifications
Presentations & Videos
Implementations of basic technologies (open source projects)
Do this on the Connected Services home page: https://at.projects.genivi.org/wiki/x/PIAVAg
- AUTOMAT Security document looks useful to read in addition to CVIM spec.
Steps/Actions (current and future to be assigned)
- AI(all): First get projects.genivi.org Wiki account if you don't have it (signup link is this one - it is part of JIRA) This is now changed and the old process should NOT be used: Please read the right column of the Wiki starting page for up to date instructions.
- Read AUTOMAT CVIM specification -- AI: Kevin -- read and summarize, present next week
- AUTOMAT Security document -- AI: Petar -- read and summarize, present next week
Future:
- Look at other AUTOMAT docs - are any other important to read?
Monday 3 June weekly call
Participants:
- Petar (Visteon)
- Christian (BMW) (partially)
- Kevin (High Mobility)
- Rafael (xapix)
- Stephen L
- Gunnar
- Philippe
- apologies: Sebastian (Bosch) vacation
Minutes
roundtable
- Rafael: xapix is a German company based in Berlin trying to create an api automotive platform www.xapix.io
- takeaway from AMM
- Kevin: liked to talk about the standardization of car data, VSS, VSS 2
- high mobility has a different perspective, establishing the framework between high mobility, the car and the cloud
- a lot of different protocols, not only VSS 2 but also several others
- discussion on the various protocols
- Gunnar: at AMM we got a presentation by cloudera on the way data are managed in an automotive api framework (big data architecture)
- Discussion followed on capturing all the different players/users of the architecture. Third-party "app" development API, is that a programming API or "only" the W3C REST-based architecture?
- High level vision on the project scope
- Gunnar: shows this slide deck
- Petar: agrees totally with slides 3&4
- one question on how to describe a behavior/service, need to represent different data types, and how to represent data that is made up of combining data from several sources.
- Gunnar: When it comes to combining signals into more refined data, the VSM project is investigating exactly that. It provides currently a python implementation. This can be useful to run on powerful IT systems, and/or be further refined to a optimized compiled code through code-generation for example.
- look at: GitHub/GENIVI/vehicle_signal_manager
- Kevin: I have a more organisational level, question, is VSS 2 more based on REST ?
- clarification brought by Gunnar on the various topics wip at W3C: VISS, VSS 2, etc.
- Clarifying VSS (Data Model @GENIVI GitHub), VISS v1 (W3C specification, which does also refer to VSS as the (minimum) data that shall be provided. VISS v2 (ongoing, in combination of augmenting VSS v2 to the needs). VISS v2 primarily focuses on adding HTTP/REST but some augmentations of the underlying features and data models happen in combination with this.
- Clarifying differences of work in data definition:
- data-model description (Exemplified by VSS - shared among implementors)
- actual data names and descriptions (Also exemplified by VSS - shared among data providers/users)
- actual data names and descriptions (E.g. proprietary extended VSS additional, e.g. OEM private)
- Notice that VSS provides two different roles - defining HOW to define data, and also WHAT data, but these can be treated separately!
- There is some talk of "other data models" in W3C, not clear if this is just other proprietary data, described in VSS-like format, or if it is described in some other format. Also not clear what those other formats might be. VSS is generally adapted to match the needs that come up.
- The fact that some companies are worried to agree that all data names/descriptions are shared tend to block the discussion. We all understand that some data definitions are unique and private and VSS v1 already described private extensions. We understand that we benefit first from a shared way to describe it, and shared tooling, second that we also can benefit from some data being shared and defined (e.g. any "developer API" *requires* that to be defined at some point), and we also agree that companies may have some own unique and private data. Key point: No part should block progress on any other part.
- Kevin asked about if W3C VISS service covers security access control features. Gunnar answered that it is specified weakly on VISS v1 (but that still allows for implementors to add it), and that in v2 we will see what makes it into the formal specification (timing, progress), but there are some discussions.
- homework
- review the draft workplan and Gunnar's slidedeck
- back to the project naming ==> homework for next week- coin other project names
- AOB
- participation to OADF open conference next week in Munich ==> check whether some of your colleagues are attending it
Monday 27 May weekly call
Participants:
- Petar (Visteon)
- Gerald (Bosch)
- Christian (BMW)
- Benjamin (BMW)
- Gunnar
- Philippe
Minutes
- roundtable
- discussion on the project name, consensus on "connected services"
- Slidedeck
- Gerald went again through the slide deck he presented at the AMM
Slide #3-4:
Gunnar: needs to identify what are the higher level protocols we use on top of radio links
discussion on the roles to play by various organizations for the definition of data containers, many organizations are currently competing on this
Gerald: where is the most interested area for the genivi group ?
Benjamin: I have multiple perspectives on those topics (slide 3), this is a highly fragmented ecosystem
Petar: IoT style communications are more interesting to me, having a connected vehicle does not mean a vehicle is always connected in my opinion
slide 6 - ISO ExVehicle
BMW: does not know who from BMW is involved
/TODO/ each participant find who in their company is involved in ISO work ExVehicle
/TODO/ Philippe contact Kevin to check whether high-mobility is implementing an alpha version of the ExVehicle concept and ask him to join the call
Gunnar: does everyone has the same understanding of what a neutral server is ?
Gunnar: as genivi we need to work on the interfaces between the various parts of the architecture
Christian: agreed, we need to define the interfaces and *do it first*
slide 6 - Sensoris
- slide 7
discussion on how CCC and W3C sync on those topics
Gunnar: in his opinion, this should not be not a constraint for our work
/TODO/ each participant find who in their company is involved in CCC work
slide 8
automat: it is likely a good idea to check this project because there are 2 OEMs involved (VW, Renault) and also there seems to be a comprehensive work on security and privacy
Philippe: IoT - FYI genivi has a liaison agreement with OCF
/TODO/ Petar look into OCF work
Petar: kuksa project is quite comprehensive
Autosar: cloud services - Benjamin discussed this with Sebastianat the AMM
/TODO/ Philippe contact Sebastian (Bosch) about the Autosar cloud services
slide 11
- Gunnar will prepare a slidedeck delivering his vision of what we should work on (in reference to the 5-level ladder shown in the panel on future vehicle electrical architectures at the AMM, look at slide #3 of future EE architectures)
Monday 20 May weekly call
Participants:
- Petar (Visteon) new participant
- Gerald (Bosch)
- Guru (Bosch)
- Gunnar
- Philippe
Minutes
- roundtable
- Petar: works with Visteon Connected Car team, he is a colleague of Giovanni (GPRO project), he is also based in Sofia, Bulgaria
- Petar is the solution architect of the team responsible for the overall architecture of the Visteon platform for connected services
- Petar has 12 year experience in connected services, for the entire stack from infrastructure to Web content end-to-end and for standard compliance
- Petar is currently more on R&D and prototyping than production
- Slidedeck
- Gerald went through the slide deck he presented at the AMM
- Gerald: session went out very well, better than expected, a lot of experience people attended, Melco (Mitsubishi Electric) which is quite involved in W3C was there and showed a lot of interest
- next steps following AMM workshop
- organization of weekly calls (instead of every other week): agreed
- /TODO/ Philippe send out an invite for the calls (starting next Monday) to the genivi-projects mailing list and the list of interested participants gathered at the AMM
- /TODO/ Philippe initiate a wiki page for the project
- feedback from today's call participants
- Petar: workplan is quite interesting, told us he identified interesting work on IoT protocol standardization, the Homie protocol specification that he came across a couple of months ago. Although it is suggested to be home IoT-oriented, it is generic enough to provide structured data interface for IoT communication, which falls in the scope of Car to Cloud.
- link: https://homieiot.github.io/
- Petar: has a question on the authorization work, is it also on the cloud side ?
- Gerald: yes, you are right, something is needed on the cloud side
- Gerald: Daniel Wilms (BMW) expressed his concern about piling up abstraction layers
- Philippe: this has been a recurrent concern from BMW for a couple of years
- Gunnar: thinks this is a no brainer
