汽车工程 ›› 2023, Vol. 45 ›› Issue (5): 768-776.doi: 10.19562/j.chinasae.qcgc.2023.05.006

所属专题: 智能网联汽车技术专题-规划&决策2023年

• • 上一篇    下一篇

无信控交叉口环境下考虑驾驶员误差的集中式轨迹规划

钱立军1,2(),陈晨1,陈健1   

  1. 1.合肥工业大学汽车与交通工程学院,合肥 230009
    2.南昌理工学院机电工程学院,南昌 330044
  • 收稿日期:2022-11-11 修回日期:2022-12-04 出版日期:2023-05-25 发布日期:2023-05-26
  • 通讯作者: 钱立军 E-mail:qianlijun66@163.com
  • 基金资助:
    国家自然科学基金面上项目(51875149)

Centralized Trajectory Planning in an Unsignalized Intersection Environment Considering Driver Error

Lijun Qian1,2(),Chen Chen1,Jian Chen1   

  1. 1.Department of Automotive and Traffic Engineering,Hefei University of Technology,Hefei 230009
    2.College of Electrical and Mechanical Engineering,Nanchang Institute of Technology,Nanchang 330044
  • Received:2022-11-11 Revised:2022-12-04 Online:2023-05-25 Published:2023-05-26
  • Contact: Lijun Qian E-mail:qianlijun66@163.com

摘要:

针对混合交通中人类驾驶汽车引起的碰撞隐患,提出一种考虑驾驶员误差的无信控交叉口集中式轨迹规划方法。首先,以最优控制框架设计了多车协同轨迹规划方法,以运动时间、燃油经济性和行车延误建立复合优化目标。其次,通过实车试验采集不同驾驶员的操作数据集,建立加速度误差的马尔科夫链误差转移概率矩阵。最后,基于车辆碰撞估计结果对可能发生事故情况进行重规划计算,并在不同自动驾驶市场渗透率工况下进行仿真验证。仿真结果表明,碰撞发生率和平均重规划次数与渗透率负相关。采用重规划方法后交叉口内的规划成功率可达90%以上,且燃油经济性等交通指标得到改善。

关键词: 混合交通, 驾驶员误差, 无信控交叉口, 轨迹规划

Abstract:

For the hidden collision risk caused by human driving vehicles (HDV) in mixed traffic, a centralized trajectory planning method for an unsignalized intersection considering driver error is proposed. Firstly, a multi-vehicle cooperative trajectory planning method is designed based on the optimal control framework, with the composite optimization objectives established based on motion time, fuel economy and driving delay. Secondly, the operating data sets of different drivers are collected through the real vehicle driving tests, then the Markov chain error transition matrix of acceleration error is established. Finally, the replanning of possible accident situation is calculated based on the results of vehicle collision estimation, and the simulation verification is carried out under different market penetration rate (MPR) conditions of autonomous vehicles. The simulation results show that the incidence of traffic collision and the average number of replanning are negatively correlated with the MPR. Moreover, the success rate of the planning at the intersection by the re-planning method can reach over 90%, with the fuel economy and other traffic indicators improved.

Key words: mixed traffic, driver error, unsignalized intersection, trajectory planning