Formal Scenario-Based Testing of Autonomous Vehicles: From Simulation to the Real World
This addresses safety verification for autonomous vehicles, which is critical for public deployment, though it appears incremental by applying formal methods to an existing testing paradigm.
The paper tackles the problem of testing autonomous vehicle safety by developing a formal scenario-based approach that spans simulation and real-world testing, demonstrating effectiveness in identifying test cases and bridging the simulation-reality gap through experiments at an industrial facility.
We present a new approach to automated scenario-based testing of the safety of autonomous vehicles, especially those using advanced artificial intelligence-based components, spanning both simulation-based evaluation as well as testing in the real world. Our approach is based on formal methods, combining formal specification of scenarios and safety properties, algorithmic test case generation using formal simulation, test case selection for track testing, executing test cases on the track, and analyzing the resulting data. Experiments with a real autonomous vehicle at an industrial testing facility support our hypotheses that (i) formal simulation can be effective at identifying test cases to run on the track, and (ii) the gap between simulated and real worlds can be systematically evaluated and bridged.