Advancing the Virtual Vehicle Fleet; E2E Vehicle Data Processing and SDV Reference Platform Integration

This presentation will provide an update on the Virtual Vehicle Fleet introduced at the AMM Spring in February. This time, we will demonstrate a virtual vehicle composed of three virtual machines: IVI, IC, and GW, incorporating the technical elements currently under consideration for the AGL SDV Reference Platform. We will delve into the complete architecture of a vehicle data processing system that spans from the vehicle to the cellular network, edge, core network, and cloud, using this virtual vehicle fleet. Additionally, we will explore the integration with a simulated 5G network (5GNW). We will also discuss the relationship between edge processing and ongoing activities at AECC.

Transcript

1. Masanori Itoh, TOYOTA MOTOR CORPORATION Advancing the Virtual Vehicle Fleet; E2E Vehicle Data Processing and SDV Reference Platform Integration
2. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. About Me Masanori Itoh  Affiliation TOYOTA MOTOR CORPORATION ✓ Digital Information and Communication Dept. InfoTech Div. ✓ Open Source Program Group (Toyota OSPO)  Works ✓ R&D for Connected Vehicle Systems • E2E Observability, Standardization, Diagnostics, … ✓ OSPO Operations  Keywords ✓ Operating System, Cloud Infrastructure, etc. ✓ https://github.com/thatsdone ✓ https://www.linkedin.com/in/masanori-itoh-6401603/ 2
3. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. Agenda 1. Virtual Vehicle Fleet – SoDeV Parity Infrastructure 2. SOVD – Beyond Diagnostics 3. AECC – Connectivity, Edge Computing & Collaboration 4. Charting the Path Forward 3
4. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. Virtual Vehicle Fleet (VVF) – SoDeV Parity Infrastructure 4
5. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. Virtual Vehicle Fleet – Previous Talk  “Virtual Vehicle Fleet in your Lab - A Use Case of AGL as a High-Level R&D Testbed Component” at AGL AMM Spring (Feb. 27, 2025) ⚫ https://sched.co/1vePR 5
6. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. VVF – Motivation and Ideas  Motivations in SDV Era ⚫ Promote Data Center side virtualization technology adoption to realistic vehicle configuration and needs for rapid development  Idea ⚫ Use the same component with SoDeV (AGL SDV Reference Platform), and build a fleet of virtual vehicles with multiple profiles of ECUs ✓ Common Components – Xen Hypervisor – Control Domain w/Zephyr ✓ Virtual Vehicles with (multiple releases of) AGL : – IVI profile – IC profile – Gateway/Telematics profile etc. ✓ Central Controller + Monitoring System ✓ Demo Dashboard 6
7. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. VVF – System Design – To Be (Demo config.)  SoDeV based Demo Configuration 7 RefHW / Raspberry Pi5 … Xen Hypervisor GPU VM#0 (dom0) Zephyr Ethernet USB#x SD GPIO VM #1 (domD) Plain Yocto or AGL compositor VM#2 (domU1) AGL IVI No HW access Xen Passthrough device tree VM#3 (domU2) AGL IC No HW access HDMI#1 HDMI#2 A Tree Swing Cartoon - https://pmac-agpc.ca/project-management-tree-swing-story VM#4 (domU3) AGL gateway (etc.) No HW access virtio (or classical xenbus based inter-vm-communication)
8. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. Virtual Vehicle #N  SoDev Inspired VVF Architecture Virtual Vehicle #2 Virtual Vehicle #1 VVF – System Design – To Be 8 ARM DataCenter Server (w/many cores + lots of memory) Xen Hypervisor GPU VM#0 (dom0) Zephyr Ethernet USB#x SD GPIO VM #1 (domD) Plain Yocto or AGL compositor VM#2 (domU1) AGL IVI No HW access Xen Passthrough (device tree?) VM#3 (domU2) AGL IC No HW access HDMI#1 HDMI#2 VM#4 (domU3) AGL gateway (etc.) No HW access virtio (or classical xenbus based inter-vm-communication) A Tree Swing Cartoon - https://pmac-agpc.ca/project-management-tree-swing-story
9. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. Virtual Vehicle #2 Virtual Vehicle #1  What is running in my Lab (at the moment) Host OS Linux (Ubuntu 24.04, arm64) VVF – System Design (current) 9 ARM DataCenter Server (w/many cores + lots of memory) KVM GPU Ethernet USB#x VM#2 AGL IVI No HW access VM#3 (domU2) AGL IC No HW access VGA Demo Dashboard (GTK+ 3.0) Windows VVF Manager (Python) dnsmasq (DHCP+DNS) libvirtd A Tree Swing Cartoon - https://pmac-agpc.ca/project-management-tree-swing-story
10. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. Virtual Vehicle #3 (Ricefish) Virtual Vehicle #2 (Salmon) VVF – Implementation (current)  KVM based Setup 10 Host OS Linux (Ubuntu 24.04, arm64) Virtual Vehicle #1 (Trout) KVM VM#1-1 AGL ivi VM#2 AGL cluster VVF Manager (Python) Windows etc. VVF Dashboard GTK+ 3.0 (Python on WSL2) Linux kernel VM#3 AGL gateway VM#3 AGL telematics libvirtd Dnsmasq (DHCP/DHCP) Should be possible to run on Linux and Mac not only Windows
11. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. (Currently) Integrated Components/Features  3GPP 5G Software Stack ⚫ Free5GC(LF) + UERANSIM ✓ https://free5gc.org/ ✓ https://github.com/aligungr/UERANSIM  Yocto layer for Cellular Network (esp. UERANSIM) ⚫ https://github.com/thatsdone/meta-cellular ⚫ Dependency: lksctp-tools included in meta-networking ⚫ Currently, installs client (UE) side binaries only 11 Follow README.md integrating to AGL (just creating a .inc file under meta-agl-devel). • Append ‘agl-cellular’ for aglsetup.sh • Add below in your local.conf IMAGE_INSTALL:append = " ueransim”
12. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. (Currently) Integrated Components/Features  Enabling BTF (BPF Trace Format) for Observability ⚫ Enables eBPF C.O.R.E (Compile Once Run Everywhere) feature ⚫ Requires ‘KERNEL_DEBUG += True’ in local.conf, and ensure to add ‘bpftool’ (which is based on libbpf not old BCC based one). ⚫ Note: kernel image/memory footprint overheads ✓ Image size +5MB, Memory footprint +? (under investigation) 12 KERNEL_DEBUG += “True“ → You would see /proc/kernel/btf/* IMAGE_INSTALL:append = " bpftool” → Add ‘bpftool’ utility $ cat /opt/yocto/configs/vvf.cfg CONFIG_DEBUG_INFO_BTF=y $ cat /opt/yocto/configs/vvf.inc FILESEXTRAPATHS:prepend:pn-linux-yocto := "/opt/yocto/configs:" FILESEXTRAPATHS:prepend:pn-linux-yocto-dev := "/opt/yocto/configs:" SRC_URI:append:pn-linux-yocto = " file://vvf.cfg " SRC_URI:append:pn-linux-yocto = " file://vvf.cfg "
13. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. SOVD – Beyond Diagnostics 13
14. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. What is SOVD  A new protocol (HTTP REST API) for vehicle diagnostics defined by ASAM ⚫ https://www.asam.net/standards/detail/sovd/ ⚫ Now under standardization process at ISO. (ISO 17978)  Excerption from the ASAM SOVD page 14 “ASAM SOVD (Service-Oriented Vehicle Diagnostics) defines an API for diagnosing and communicating with software-based vehicles. It is a flexible standard that provides a uniform access to the diagnostic content of HPCs and their related applications, as well as of classic ECUs. ” Diagram excerpted from the above ASAM SOVD page.
15. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. SOVD – Use Cases  Use Cases ⚫ Key differences to classical diagnostics described in ISO 17978-2 ‘Road vehicles — Service-oriented vehicle diagnostics (SOVD) — Part 2: Use cases definition 15 1. Discover SOVD Server(s) 2. Discover Entities 3. Authenticate and Authorise SOVD Clients 4. Read Fault(s) and/or its Details from an Entity 5. Delete Fault(s) of an Entity 6. Read Data Resources 7. Write Data Items 8. Control Operations 9. Read and Write Configurations 10.Control Software Updates 11.Handle Bulk Data 12.Control Logging and Retrieve Logs 13.Control Communication Logging 14.Manage and Execute Scripts 15.Manage Triggers and subscriptions 16.Query an Online Capability Description 17.Provide an Offline Capability Description Dynamic discovery including Software OTA : Over-The-Air (software update) Runtime is OEM specific, but SOVD can use script based in- vehicle generic processing
16. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. SOVD based Diagnostics and Control 16 TCU ECU UDS-GW Classic ECU#1 WiFi OBD2 Remote Diagnostics / Data Collection Co-Located Technician (via Direct WiFi) Cellular NW Co-Located #2 WiFi ECU (HPC) Classic ECU#2 SOVD server #1 CAN Ethernet Remote Cellular Tester Co-Located Technician (via OBD2) Internet Co-Located #1 Co-Located Technician (through the Internet) Co-Located #3
17. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. SOVD – REST API Examples  REST API ⚫ Actual REST API operation examples defined in ISO 17978-3 ‘Road vehicles — Service-oriented vehicle diagnostics (SOVD) — Part 3: Application programming interface (API)’ 17 Example 2 : Get Application (Advanced Lane Keeping) information GET /apps/AdvancedLaneKeeping HTTP/1.1 Example 1： Retrieve PowerSteering status information GET /components/PowerSteering HTTP/1.1 Example 3 : Issue script execution POST /components/DrivingComputer/scripts/CollectAppMetrics/executions HTTP/1.1 (*) Parameters are specified via the request body
18. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. OpenSOVD – an Open Source Implementation  Eclipse OpenSOVD ⚫ Mercedes lead initiative to implement an open source SOVD ecosystem ✓ https://projects.eclipse.org/projects/automotive.opensovd ✓ https://github.com/eclipse-opensovd  Kicked-off on July 2, 2025 ⚫ More than 90 participants! ✓ https://www.youtube.com/watch?v=VnMauUXT2cI ✓ https://drive.google.com/file/d/1vGNs7pPWI0RLHT3dWBQt24PV7vZa7yTQ/view 18
19. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. SOVD – Beyond Diagnostics 19
20. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. AECC – Connectivity, Edge Computing & Collaboration AECC : Automotive Edge Computing Consortium (https://aecc.org) ⚫ An Open Collaborative Activity for promoting Edge Computing focusing on Automotive Use Cases, with POCs 20 Vehicle System Connected Car Vehicle System Connected Car Vehicle System Connected Car Vehicle System Connected Car Cellular Network Edge Server Pre- processing Cloud Edge Server Pre- processing Moving logs Biometric Sensors Cameras Cruising Data Sensor Data Control Data Intelligent Driving Enterprise Network CAN
21. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. AECC Activities  White Papers ⚫ https://aecc.org/resources/publications/  POCs ⚫ https://aecc.org/proof-of-concepts/ ⚫ “Optimal Edge Selection for Realizing Digital Twins and Green Energy Utilization” (POC10) ✓ Shooting for Carbon Neutral by selecting optimal Compute Resources spread across Regions by choosing optimal routes/Edges  SIGs based on particular solution areas (now under forming process) ⚫ (1) Data-first SIG ✓ Starting from V2V enabling more flexible/reliable vehicle data utilization ⚫ (2) Telco API SIG ✓ CAMARA API + Remote SIM Provisioning ⚫ (3) Green Computing SIG ✓ How to utilize green energy distributing workload by optimizing edge/route selection 21
22. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. Charting the Path Forward 22
23. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. Integrated POC using SoDeV and more at AECC 23 SoDeV VVF 5GS Stack SOVD POC w/Vehicle Data Utilization and more… (e.g. Kubernetes) + + + +
24. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. A Big-Picture of the POC A POC System of Connected Vehicles Service Data Processing gNB#1 W A N Location #1 (e.g. Tokyo) Location #2 (e.g. Osaka) Public Cloud #1 CAN Camera DynamicMap Others… LB Ops Subsystem Pseudo Vehicle Virtual Vehicle (SoDeV) Edge#1(Region #2) Auth GW Dispatcher Offload Process Slice#1 (Lat.) Auth GW For Fallback W A N Data Accumulation Edge#2(Region #2) Auth GW Dispatcher Offload Process UE#1 Slice#2 (B/W) UPF#2 UPF#1 Dedicated Line Virtual Vehicle (SoDeV) UE#1 Slice#1 (Lat.) Slice#2 (B/W) OCI /etc. Public Cloud#2 UPF#2 UPF#1 Fail Over gNB#2 gNB#2 gNB#1 Orchestra tion gNB#1 Dedicated Line/VPN Dedicated Line/VPN Dedicated Line/VPN 24
25. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. Ideas for the POC(s) at AECC Random Thoughts (under discussion) ⚫ Use Cases ✓Offer location aware tailored information service to drivers ✓Digital Twin at transportation hubs/intersections ✓etc. ⚫ Enabling Solutions ✓On-demand vehicle data retrieval via next-gen. diagnostics ✓Hierarchical Agentic AI ✓etc. 25
26. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. Summary 26
27. Copyright © 2025 TOYOTA MOTOR CORPORATION All rights reserved. Summary Realistic Virtual Vehicle Fleet based on SoDeV Beyond Diagnostics by SOVD POCs integrating SOVD, 5G stack and more working with external communities (starting from AECC) 27