A Risk-Index Based Sampling Method to Generate Scenarios for the Evaluation of Automated Driving Vehicle Safety

Abstract

This paper presents a novel framework to generate scenarios for the evaluation of autonomous vehicle safety. The test scenarios are generated by sampling parameters from a probabilistic model based on naturalistic driving data, which consist of vehicle kinetic information and a traffic risk index. The proposed framework also allows to estimate the comprehensiveness of the test scenario with respect to the naturalistic driving dataset. Further, it flexibly adopts arbitrary definitions of the relationships between kinetic parameters and the risk index. To evaluate the effectiveness of the proposed method, we conducted experiments by using simulated data and naturalistic driving data. By using the proposed method for safety validation enables that the automatic generation of test scenarios based on driving accident risk indices.