A Formal Safety Characterization of Advanced Driver Assist Systems in the Car-Following Regime with Scenario-Sampling
This work addresses safety evaluation for ADAS in car-following scenarios, offering a more reliable method for developers and regulators, though it builds incrementally on prior safety quantification research.
The paper tackles the problem of evaluating safety in car-following systems for ADAS, which often rely on biased metrics or require extensive testing, by proposing a scenario-based framework that guarantees unbiased and efficient sampling, demonstrating its performance on widely used modules and a commercial driving stack.
The capability to follow a lead-vehicle and avoid rear-end collisions is one of the most important functionalities for human drivers and various Advanced Driver Assist Systems (ADAS). Existing safety performance justification of the car-following systems either relies on simple concrete scenarios with biased surrogate metrics or requires a significantly long driving distance for risk observation and inference. In this paper, we propose a guaranteed unbiased and sampling efficient scenario-based safety evaluation framework inspired by the previous work on $εδ$-almost safe set quantification. The proposal characterizes the complete safety performance of the test subject in the car-following regime. The performance of the proposed method is also demonstrated in challenging cases including some widely adopted car-following decision-making modules and the commercially available Openpilot driving stack by CommaAI.