Scholarlink

Efficient and reliable motion planning system is critical in changing environments for autonomous driving. In this paper, we present a motion planning algorithm for dynamic scenarios through Gaussian process(GP) path planner and trajectory predictor. Firstly we plan a feasible path with GP planner. Then, the predictor generates several possible trajectories of other participants and we use a S-T graph speed planner to produce the speed profile with predicted results. Finally, simulation results demonstrate that our algorithm can improve the success rate of random driving tasks compared to the commonly used constant velocity assumption.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Motion Planning Framework with Learning Based Trajectory Prediction in Self Driving

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Feiyu Bian,
  • Xing Liu,
  • Yizhai Zhang,
  • Zhiqiang Ma,
  • Ganghui Shen,
  • Panfeng Huang

摘要

Efficient and reliable motion planning system is critical in changing environments for autonomous driving. In this paper, we present a motion planning algorithm for dynamic scenarios through Gaussian process(GP) path planner and trajectory predictor. Firstly we plan a feasible path with GP planner. Then, the predictor generates several possible trajectories of other participants and we use a S-T graph speed planner to produce the speed profile with predicted results. Finally, simulation results demonstrate that our algorithm can improve the success rate of random driving tasks compared to the commonly used constant velocity assumption.