Human Behavior-Informed Crash Scenario Generation with Real-World Crash Priors for Autonomous Vehicle Safety Evaluation
CrashSim is presented, a human behavior-informed crash scenario generation framework that uses real-world crash priors to guide generative multi-agent traffic simulation for more reliable AV safety evaluation and uses nuCrash dataset, containing over 4,000 crash and near-crash scenarios.
Ming-Xing Peng, Xu-Sen Guo, Long Chen et al.
· 0 citations