基于轨迹预测模型的仿真车辆轨迹生成算法  

Trajectory generation algorithm for simulated vehicles based on trajectory prediction models

作　　者：王振宇 余卓平[1] 田炜 熊璐[1] 李拙人 WANG Zhenyu;YU Zhuoping;TIAN Wei;XIONG Lu;LI Zhuoren(School of Automotive Studies,Tongji University,Shanghai 201800,China)

机构地区：[1]同济大学汽车学院,上海201800

出　　处：《汽车安全与节能学报》2025年第2期286-293,共8页Journal of Automotive Safety and Energy

基　　金：国家重点研发计划项目(2022YFE0117100)。

摘　　要：为了提升自动驾驶数字仿真场景生成算法中背景交互车辆行驶轨迹的整体真实性,该研究从微观和宏观2个层面切入,首先基于自然驾驶数据训练车辆轨迹预测模型,利用模型预测轨迹更加贴近真实场景车辆轨迹的特点,将其作为仿真场景中背景车辆的人工智能(AI)驾驶员模型,提升仿真车辆轨迹交互的微观真实性;在此基础上,设计轨迹特征参数统计分布差异度量方法和针对性的优化算法,从预测模型输出的多条多模态预测轨迹中重新选取单条概率最高的最优轨迹,使其作为仿真车辆的行驶轨迹,进一步提升生成轨迹特征参数统计分布的宏观真实性。结果表明:基于该研究提出的度量指标,优化后算法输出的仿真轨迹与真实轨迹的分布差异降低了56.29%,有效提升了仿真场景中背景车行驶轨迹的宏观真实性。To enhance the overall realism of background interactive vehicle trajectories in digital simulation scenarios for autonomous driving,this study approached the problem from both microscopic and macroscopic perspectives.Firstly,vehicle trajectory prediction models were trained on naturalistic driving data.Leveraging the characteristic that model-predicted trajectories more closely resembled real-world vehicle trajectories,the prediction served as the artificial intelligence(AI)driver model for background vehicles in simulation environments,improving the microscopic realism of simulated vehicle trajectory interactions.Building on this foundation,a measurement method for trajectory feature parameter statistical distribution differences and a corresponding optimization algorithm were designed,to re-select a single trajectory with the highest probability from multiple multi-modal prediction outputs,as the final driving trajectory for simulated vehicles,further enhancing the macroscopic realism of the generated trajectory feature parameter statistical distribution.The results show that,based on the proposed measurement metrics,the distribution difference between optimized simulated trajectories and real trajectories is reduced by 56.29%compared to pre-optimization,effectively enhancing the realism of background vehicle trajectories in simulation scenarios.

关 键 词：多模态轨迹预测 轨迹快照 轨迹特征向量聚类 Kullback-Leibler(KL)散度 Bayes优化 

分 类 号：U492.84[交通运输工程—交通运输规划与管理]