基于虚拟触角的自主车辆运动规划方法

在对自主车辆进行运动学分析的基础上，结合反应式行为规划方法和昆虫利用触角进行避障行走的思想，提出了一种使用圆弧形虚拟触角进行运动规划算法，并规划出在各圆弧轨迹片段车辆前轮摆角的参考值。仿真和实验结果表明：通过设置适当的初始参数，该算法能够使自主车辆在行驶过程中完全避开障碍物到达目标点，且轨迹光滑，满足车辆的运动学约束条件。该算法计算简洁、实时性强，适用于底层的实时控制，为自主车辆运动规划提供了一种简便、有效的策略。

A Motion Planning Method for Autonomous Land Vehicles Based on Virtual Tentacles

Based on the kinematics analysis of autonomous land vehicles,in cooperation with the reactive action planning method and the insect′s obstacle avoidance by tentacles,a novel motion planning algorithm using circular-arc virtual tentacles is proposed and the reference front wheel angle of the arc trajectory fragment is given.The simulation and experimental results show that if the proper initial parameters are set,the virtual tentacles algorithm can enable the autonomous vehicle to avoid an obstacle with smooth track and meet the kinematics constraint.It is proved that the algorithm is concise and fit for the real-time control,providing a simple and effective strategy for the motion planning of autonomous vehicles.