Test Concrete Scenarios Extraction of Lane-Changing Scenarios Based on China-FOT Naturalistic Driving Data

On account of the insufficient lane-changing scenario test cases and the inability to conduct graded evaluation testing in current autonomous driving system field testing, this paper proposed an approach that combined data-driven and knowledge-driven methods to extract lane-changing test concrete scenarios with graded risk levels for field testing. Firstly, an analysis of the potentially hazardous areas in lane-changing scenarios was conducted to derive key functional lane-changing scenarios. Three typical key functional lane-changing scenarios were selected, namely, lane-changing with a preceding vehicle braking, lane-changing with a preceding vehicle in the same direction, and lane-changing with a rear cruising vehicle in the adjacent lane, and their corresponding safety goals were respectively analyzed. Secondly, the GAMAB criterion was introduced as an evaluation standard for autonomous driving systems. By utilizing lane-changing scenario data selected from the China-FOT naturalistic driving data, a scenario risk classification model and a model for excellent driver response performance in lane-changing scenarios were established. Finally, concrete scenarios corresponding to different risk levels for each type of lane-changing scenario were extracted through simulation. Test concrete cases for field testing were selected at the risk boundaries based on the characteristics of China-FOT naturalistic driving data. The results demonstrated that the proposed approach was capable of effectively extracting 701 high-risk scenarios and 446 medium-risk scenarios from a pool of 9000 concrete scenarios based on key functional lane-changing scenarios. Furthermore, representative lane-changing test concrete cases can be selected from the risk boundaries. This approach enabled a graded evaluation of the lane-changing capability of the autonomous driving system.