Safety Analysis of Autonomous Driving Systems Based on Model Learning
This work addresses safety verification for autonomous driving systems, which is critical for public safety, though it appears incremental as it builds on existing model learning and verification techniques.
The researchers tackled the problem of verifying safety in autonomous driving systems by building surrogate models that quantitatively depict system behavior in traffic scenarios, achieving probabilistic safety guarantees for the original systems. They demonstrated their approach by evaluating safety properties on state-of-the-art autonomous driving systems across various simulated traffic scenarios.
We present a practical verification method for safety analysis of the autonomous driving system (ADS). The main idea is to build a surrogate model that quantitatively depicts the behaviour of an ADS in the specified traffic scenario. The safety properties proved in the resulting surrogate model apply to the original ADS with a probabilistic guarantee. Furthermore, we explore the safe and the unsafe parameter space of the traffic scenario for driving hazards. We demonstrate the utility of the proposed approach by evaluating safety properties on the state-of-the-art ADS in literature, with a variety of simulated traffic scenarios.