Virtual Safety Assessment of Autonomous Vehicle Functions

Automated driving systems tend to be more prominent and sophisticated in the nearest future. The functional safety assessment for these systems becomes an urgent necessity for the transition to full autonomy. Testing these functions consisting of decision and control algorithms, in a unified manner, is a challenging task. This task requires the analysis of complex operational profiles; such as routing, intersection management, and collision prediction, in an environment where multiple vehicles emerge from different positions, traveling at different speeds. There is a need for a comprehensive traffic simulation framework which does not only model the functionality of the vehicles but also the interactions between them. Brute force testing is not an option where billions of hours of test drive are necessary for validating the safety of autonomous driving functions(ADF).

Our Goals:

  • Benchmark ADF/ADAS by generating safety/performance reports.
  • Run PC-based simulations to validate autonomous vehicle ECUs (Electronic Control Units) virtually before a prototype is available.
  • Detect bugs in early phases and support agile development iterations.
  • Reduce production costs and time to market.

Our Solutions:

  • Model-based design of autonomous and connected vehicles, development of MOBATSim, versatile, standard and flexible: 100% MATLAB Simulink.
  • Methodology: Determination critical and rare traffic situations, Simulation-based fault injection and driving scenario-based testing.
  • Prospective Safety analysis: Report generation for Virtual Validation, guided simulation methods for efficient testing and focusing computing power.
MOBATSim Virtual Safety Assessment