首页|基于迭代二次型调节器的自动驾驶行为规划应用研究

基于迭代二次型调节器的自动驾驶行为规划应用研究

扫码查看
针对自动驾驶行为规划问题提出了 一种基于迭代二次型调节器(iLQR)的交通流仿真算法.首先,对于需要行为规划的强交互场景(路口,换道等),找到交互影响最大的交通参与者;然后,进行策略采样,对每种策略进行双车交通流仿真,得到每种策略下双车的轨迹;最后通过对各个策略下生成的轨迹进行评估,得到最合理的交互策略.实验结果表明,基于迭代二次型调节器的最优控制算法可以显著提升求解行为规划问题的效果和降低耗时.
Application Research on Autonomous Driving Behavior Planning Based on Iterative Quadratic Regulator
A traffic flow simulation algorithm based on iterative quadratic regulator(iLQR)is proposed for autonomous driving behavior planning problems.Firstly,for strong interaction scenarios that require be-havior planning(intersections,lane changes,etc.),identify the traffic participants with the greatest inter-action impact;Then,strategy sampling is conducted to simulate the dual vehicle traffic flow for each strate-gy,and the trajectories of the dual vehicles under each strategy are obtained;Finally,by evaluating the trajectories generated under each strategy,the most reasonable interaction strategy is obtained.The experi-mental results show that the optimal control algorithm based on iterative quadratic regulators can signifi-cantly improve the effectiveness of solving behavior planning problems and reduce time consumption.

decision planningbehavior planningiLQRtraffic flow simulation

藏中阳、黄宏成

展开 >

上海交通大学机械与动力工程学院,上海 200240

决策规划 行为规划 iLQR 交通流仿真

2024

传动技术
上海交通大学

传动技术

影响因子:0.197
ISSN:1006-8244
年,卷(期):2024.38(1)
  • 7