10 th AUTOSAR Open Conference

Transcription

1 10 th AUTOSAR Open Conference Pierre-Antoine Bernard OpenSynergy GmbH Cornel Izbasa OpenSynergy GmbH Virtualization Solutions for the AUTOSAR Classic and Adaptive Platforms AUTOSAR Nov-2017

2 OpenSynergy GmbH Global provider of software and services for automotive embedded systems to Tier-1 suppliers and OEMs Development Member Headquarters in Berlin (Germany), further locations in Munich, San Diego and Salt Lake City Platform virtualization software (COQOS, COQOS micro) as part of the product portfolio Focus on cockpit controller, connectivity (smart antenna) and ADAS systems AUTOSAR Nov

3 Outline Hardware Consolidation Software Consolidation AI for Autonomous Driving and Virtualization Semantic Abstraction and Virtualization Informational ADAS and Virtualization Hypervisor Inter-VM Communication AUTOSAR Nov

4 Hardware Consolidation System-on-Chip (SoC) CPU Cores CPU Cores Video Decoder GPU IPU DSP Ethernet AVB CAN FD Flexray SPI USB Audio IF Display IF Camera IF Hardware Consolidation Silicon manufacturers tend to intensify the integration of IP blocks inside SoCs: Clusters of powerful homogeneous or heterogeneous CPU cores I/O interfaces (Ethernet AVB, CAN FD ) Acceleration blocks (video decoder, DSP ) AUTOSAR Nov

5 Software Consolidation Why virtualization? Heterogeneous SW platforms onto a single SoC Efficient use of highly integrated SoC Extra SW layer between HW and OSs Reduced attack surface (limited API) Overall cost reduction Modular software update Instrument Cluster SW Head Unit SW ADAS SW Connectivity SW Virtualization Layer System-on-Chip Benefits Safety (real-time, freedom from interference) Enhanced security Use of automotive standards Low impact on performance AUTOSAR Nov

6 Artificial Intelligence for Autonomous Driving I m a Cartesian Dualist! Symbolic Putative characteristics of cognition Information processing metaphor Subsymbolic Inspired by neurobiology How cognition emerges from neurobiology Expert systems Artificial Neural Networks Deep Neural Networks (DNN) Based on The Mind-Body Problem from [8], using [1] for IP reasons. AUTOSAR Nov

7 Deep Learning Architectures for Autonomous Driving End-to-End Input Sensor Data DNN Output Vehicle Control Semantic Abstraction Input Sensor Data VD PD LD Control DNN / algo Output Vehicle Control Diagrams based on [5] Semantic Abstraction is superior to End-to-End architecture, for a given amount of training data. [4] demonstrates cases in which End-to-End requires exponentially greater number of examples. AUTOSAR Nov

8 Semantic Abstraction enables Virtualization Semantic Abstraction enables running the different neural network and algorithmic components in different VMs, ensuring freedom from interference. ADAS/AD SW Instrument Cluster SW Head Unit SW Connectivity SW VD 1 Adaptive AR VD2 Adaptive AR PD Adaptive AR LD Adaptive AR Control Logic Adaptive AR ASIL B ASIL D Virtualization Layer System-on-Chip AUTOSAR Nov

9 Efficient Informational ADAS through ADAS/IC/IVI Consolidation Informational ADAS, helpful for societal, individual and even legal acceptance of AD [7], is easier and more efficiently realized on a system that integrates ADAS, IC and IVI functionality in separate VMs with mixed criticality. Decision explanation challenging for Subsymbolic AI, but see [6] for related work. Examples of salient objects, inspired by [6], based on [3] for IP reasons. AUTOSAR Nov

10 Hypervisor memory management Guest OS exception handling Hypervisor Multi-Core HW I/O resource management Application VM memory management exception handling interrupt handling interrupt handling 2 nd level MMU 1 st level MMU INTC vintc Non- Privileged System Call Privileged Hyper Call Hyper Privileged Type-1 hypervisor CPU abstraction (multi-core capable) Containment and scheduling of VMs Highest CPU operation mode Non-monolithic architecture Access control to I/O Low footprint AUTOSAR Nov

11 Inter-VM Communication Application Kernel API Monolithic Kernel VM Driver VM Application Driver A Hypervisor provides the low level communication mechanisms (multi-core capable) to set the foundation for high level communication frameworks (virtual Ethernet for example): Shared memory Hypercall Inter-VM Communication Hypervisor The communication properties may be similar to the AUTOSAR OS IOC (Inter-OS- Application Communicator): Event vs Last-is-Best semantic Triggering on data reception AUTOSAR Nov

12 Virtual Ethernet Virtual Ethernet Switch VM Application Kernel Virtual Ethernet Driver Service VM Virtual Ethernet Switch VM Application Kernel Virtual Ethernet Driver Virtual Ethernet enables the use (or reuse) of an IP communication stack. Applications are able to perform high level inter-vm communication (BSD socket as kernel interface). Inter-VM Communication Hypervisor AUTOSAR Nov

13 Use Case: Multi-AUTOSAR Classic AUTOSAR vecu Classic AUTOSAR vecu Service VM Hypervisor Adaptive AUTOSAR vecu Classic AUTOSAR Time and space partitioning Cluster of VMs to implement multiple vecus VM SW update granularity Different SW suppliers Easier development across multiple SW suppliers AUTOSAR Nov

14 Classic AUTOSAR: Inter-VM Communication Classic AUTOSAR Classic AUTOSAR Metadata SWC SWC SWC Description Gen CDD Description RTE RTE CDD Code Gen ECU Configuration Inter-VM Communication CDD Inter-VM Communication CDD RTE Code Gen Inter-VM Communication Hypervisor Data proxying overhead (CDD) Dedicated tooling required Not standardized AUTOSAR Nov

15 Classic AUTOSAR: Standardized Inter-VM Communication Standardized Module (Proposal): Signal Interface to BSW and RTE Last-is-Best and Event semantics Polling or (virtual) interrupt driven Similar to OS IOC May be an extension of the IOC submodule instead of a separate module Could pave the road for a distributed BSW across VMs Classic AUTOSAR SWC RTE VirtCom SWC Classic AUTOSAR RTE IOC OS SWC Inter-VM Communication Hypervisor AUTOSAR Nov

16 Classic AUTOSAR: Service-Oriented Inter-VM Communication Classic AUTOSAR Xf LdCom PduR SWC RTE VirtCom BswM SD (Inter-VM SD) Adaptive AUTOSAR Classic AUTOSAR Xf LdCom PduR SWC RTE SoAd TcpIp EthIf BswM SD (SOME/IP SD) Eth (virtual) Possible Solution: PDU interface towards PduR and SD Inter-VM SD specific protocol No overhead of TCP, IP and Ethernet (OSI Layer 2) Inter-VM Communication Hypervisor AUTOSAR Nov

17 Adaptive AR: Service-Oriented Inter-VM Communication Adaptive Application Adaptive Application ara::com Inter-ECU binding SOME/IP, DDS, etc. veth ara::com ara::com inter-vm binding Object Graph Inter-VM Communication Virtualization Layer System-on-Chip ara::rest HTTP/JSON veth Adaptive AR Platform Instance Adaptive AR Platform Instance Classic AR Platform Instance Consolidation Adaptive AR Platform Instance Classic AR Platform Instance Ethernet AUTOSAR Nov

18 Standard interface for ara::com bindings on Adaptive AUTOSAR? Adaptive Application Adaptive Application ara::com Standard interface? Standard interface? Standard interface? ara::com Inter-ECU binding(s) SOME/IP, DDS, etc. ara::com inter-vm binding ara::com local IPC binding Adaptive AUTOSAR Platform Instance Allows integration of binaries from Adaptive Application providers onto a project s specific communication binding(s). Ease testing of application code using mock bindings. AUTOSAR Nov

19 Hypervisor Concept for AUTOSAR OpenSynergy aims to bring the hypervisor concept into AUTOSAR through the creation of a CONC document and related artifacts such as: Methodology and metamodel Software modules and APIs OpenSynergy welcomes the cooperation with other AUTOSAR members on this topic. AUTOSAR Nov

20 References 1. Brain image: 2. Data is the New Oil in the Future of Automated Driving, Brian Krzanich, Intel, 3. Road image: 4. On the Sample Complexity of End-to-end Training vs. Semantic Abstraction Training, S. Shalev-Shwartz, A. Shashua, Mobileye, 5. Autonomous Car - what goes into sensing for autonomous driving?, presentation by A. Shashua, Mobileye, I?t= Explaining How a Deep Neural Network Trained with End-to-End Learning Steers a Car, M. Bojarski et al., NVIDIA Corporation, 7. European Union regulations on algorithmic decision-making and a right to explanation, B. Goodman, S. Flaxman, 8. The Symbolic vs Subsymbolic Debate, H. Bowman, CCNS, Kent, AUTOSAR Nov

21 10 th AUTOSAR Open Conference Thank you for your attention! Pierre-Antoine Bernard OpenSynergy GmbH Rotherstraße Berlin / Germany t pierre-antoine.bernard@opensynergy.com Cornel Izbasa OpenSynergy GmbH Rotherstraße Berlin / Germany t cornel.izbasa@opensynergy.com - AUTOSAR - page 21