A Versatile Approach to Evaluating and Testing Automated Vehicles based on Kernel Methods
This work addresses the challenge of safety validation for automated vehicles, offering an incremental improvement over existing accelerated evaluation methods by automating distribution construction.
The paper tackles the problem of efficiently testing rare but critical failures in automated vehicles by proposing a versatile approach using kernel methods to construct sampling distributions, resulting in robust identification of rare failures and significant reduction in evaluation time.
Evaluation and validation of complicated control systems are crucial to guarantee usability and safety. Usually, failure happens in some very rarely encountered situations, but once triggered, the consequence is disastrous. Accelerated Evaluation is a methodology that efficiently tests those rarely-occurring yet critical failures via smartly-sampled test cases. The distribution used in sampling is pivotal to the performance of the method, but building a suitable distribution requires case-by-case analysis. This paper proposes a versatile approach for constructing sampling distribution using kernel method. The approach uses statistical learning tools to approximate the critical event sets and constructs distributions based on the unique properties of Gaussian distributions. We applied the method to evaluate the automated vehicles. Numerical experiments show proposed approach can robustly identify the rare failures and significantly reduce the evaluation time.