基于改进混合A^(*)算法的智能汽车路径规划  

Intelligent Vehicle Path Planning Based on Improved Hybrid A^(*) Algorithm

作　　者：潘锐 周红妮 郭昊 Pan Rui;Zhou Hongni;Guo Hao(Hubei Key Laboratory of Automotive Power Train and Electronic Control,Shiyan 442002,China)

机构地区：[1]汽车动力传动与电子控制湖北省重点实验室,湖北十堰442002

出　　处：《湖北汽车工业学院学报》2025年第1期9-14,共6页Journal of Hubei University Of Automotive Technology

基　　金：湖北省教育厅科研计划优秀中青年项目(Q201918050);湖北省创新基金(2015XTZX0414)。

摘　　要：针对混合A^(*)算法在非结构化道路路径规划效率低、规划路径不够平滑等问题,提出改进混合A^(*)算法,在实际代价中加入对方向切换、航向角的惩罚,采用指数加权法对启发函数进行加权;采用变步长方式拓展节点,并对终点区域路径进行RS曲线拟合;利用梯度下降法对路径进行平滑处理和避障优化。通过构建复杂场景和泊车场景进行路径规划仿真,结果表明:改进后的混合A^(*)算法相比于传统算法,在路径长度、拓展节点、搜索时间以及曲率等方面均有明显的优势,可以有效解决非结构化道路的路径规划问题。An improved hybrid A^(*) algorithm was proposed to address the issues of low efficiency in unstructured road path planning and insufficient smoothness of planned paths by the original algorithm.The penalties for direction switching and heading angle were added to the actual cost,and the exponential weighting method was adopted to weight the heuristic function.Nodes were expanded using a variable step size method,and RS curve fitting was applied to the path in the end area.The gradient descent method was used to smooth the path and optimize the obstacle avoidance.The path planning simulation was carried out by constructing complex scenarios and parking scenes.The results show that the improved hybrid A^(*) algorithm has obvious advantages in path length,expansion nodes,search time,and curvature compared with the traditional algorithm,able to effectively address the path planning challenges in unstructured roads.

关 键 词：混合A^(*)算法 非结构化道路 启发函数 梯度 

分 类 号：U463.6[机械工程—车辆工程]