Physics Based Sensor simulation Jordan Gorrochotegui - Product Manager Software and Services Mike Phillips Software Engineer Restricted Siemens AG 2017 Realize innovation.
Siemens offers solutions across all automotive mega trends 7 1 2 6 3 5 4 Autonomous Vehicles Electrified Vehicles Mobility Connected Vehicles Autonomous driving & driver assist systems Electric vehicles & supporting technology Smart fleets & multi-modal transport Connecting car, driver and infrastructure Engineering the next product not just the best product for the future
Validation and Verification framework for AVs Requirements 1M 10M scenarios Multiple variants V&V environments MiL / SiL / Cluster Requirements & system architectures Digital Twin World Digital Twin Vehicle 1k 10k scenarios HiL / DiL / ViL Simulation definition 10 100 scenarios Proving ground / field test Certification - Homologation Real world Design adaptations (HW/SW) Vehicle under development Page 3
Validation and Verification framework for AVs Requirements 1M 10M scenarios Multiple variants V&V environments MiL / SiL / Cluster Requirements & system architectures Digital Twin World Digital Twin Vehicle 1k 10k scenarios HiL / DiL / ViL Simulation definition 10 100 scenarios Proving ground / field test Certification - Homologation Real world Design adaptations (HW/SW) Vehicle under development Page 4
Example #1: MiL / SiL / Cluster V&V Environments MiL / SiL / Cluster Run massive amounts of Prescan scenarios for Automated Vehicle development and optimization 1k 10k scenarios HiL / DiL / ViL 10 100 scenarios Laboratories / proving grounds / public roads Design space exploration using large scenario databases Virtual development, verification and robustness testing Optimized automated vehicle designs Page 5
Example #2: HiL testing of central AD processing unit V&V Environments MiL / SiL / Cluster PreScan synthetic sensor data injection for virtual validation of central AD processing units Automated Driving processing unit 1k 10k scenarios HiL / DiL / ViL Automated Driving system validation 10 100 scenarios Laboratories / proving grounds / public roads Page 6 Training and evaluating Deep Neural Networks (DNNs) Virtual validation of automated driving processing units Accelerated automated vehicle development
Example #2: HiL testing of central AD processing unit Free space detection on Nvidia Drive PX2 Page 7
Example #2: HiL testing of central AD processing unit Object detection on Mentor DRS360 Page 8
PreScan HIL application examples V&V Environments MiL / SiL / Cluster 1k 10k scenarios HiL / DiL / ViL 10 100 scenarios Laboratories / proving grounds / public roads Page 9
Example #3: Automated Driving physical validation V&V Environments MiL / SiL / Cluster TASS International Services and Siemens Testing Solutions for physical validation of automated and connected driving technology 1k 10k scenarios HiL / DiL / ViL 10 100 scenarios Page 10 Laboratories / proving grounds / public roads Physical verification & validation services Certification of automated and connected systems Design consultancy for next-generation AD test facilities
Development of automated vehicles requires realistic sensor data Algorithm development Sensor development DNN Training Validation Page 11
Two sources of sensor data + Real data - Expensive - Time consuming - Not (easily) repeatable - Requires a physical sensor - Must be annotated - Open loop (recorded)? Realistic data + Inexpensive + Fast to acquire + Perfect repeatability + Physical sensor not needed + Annotation is free + Can be closed loop Recorded Simulated Page 12
Camera & Ground Truth Radar & V2X Lidar & Point Cloud Page 13
We use simulation to compute real world effects on actual sensors Physical artefacts are faithfully reproduced Raw signal is computed Output verified & validated Distortions Multi-bounce Time effects Weather Lidar full-waveform signal Radar channel response Raw camera images Verify against real sensors Validate for specific use cases Page 14
Color filter array Simulation Engine Beam cross section MTF Three key components are needed Physical Device World Page 15
Physics Based Lidar Simulation Page 16
Our solutions support autonomous vehicle development needs Across all engineering domains Ensuring digital continuity, multi-domain traceability and functional safety of autonomous systems Page 17