用于无人车路径规划的改进A*算法  

An improved A*algorithm for unmanned vehicle path planning

作　　者：田海波[1] 李陆军 畅科剑 高超 TIAN Haibo;LI Lujun;CHANG Kejian;GAO Chao(School of Mechanical Engineering,Xi’an University of Science and Technology,Xi’an 710054,China)

机构地区：[1]西安科技大学机械工程学院,西安710054

出　　处：《现代制造工程》2021年第11期63-68,92,共7页Modern Manufacturing Engineering

基　　金：国家自然科学基金项目(51705412,51975468,52174149);陕西省自然科学基础研究计划项目(2015JM5235);同济大学嵌入式系统与服务计算教育部重点实验室开放课题基金项目(ESSCKF2015-04)。

摘　　要：针对传统A*算法规划的路径存在路径长度较长、拐点数量较多和转折角度较大等问题,提出了一种用于无人车路径规划的改进A*算法。首先在传统A*算法的基础上考虑转弯成本并融入预判断规划策略,规划出一条初步路径;然后采用冗余拐点剔除策略优化初步路径,获得一种更优的无人车行驶路径。对传统A*算法、考虑转弯成本并融入预判断规划策略的A*算法和改进A*算法进行仿真对比。结果表明:改进A*算法规划出的路径长度更短,拐点数量更少,转折角度更小,路径更加平滑。An improved A*algorithm was proposed to solve the problems that path length is long,inflection point number and turning angle are large.First,on the basis of the traditional A*algorithm,the cost of turning was considered and the path planning strategy based on pre-judgment was integrated to plan a preliminary path.Then,the redundant inflection point elimination strategy was adopted to optimize the preliminary path so as to obtain a better driving path for unmanned vehicle.Simulation was carried out on the traditional A*algorithm,the A*algorithm considering the turning cost and incorporating the pre-judged planning strategy,and the improved A*algorithm.And the results show that the improved A*algorithm has shorter path length,fewer inflection point,smaller turning angle and smoother path.

关 键 词：A*算法 路径规划 无人车 估价函数 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]