摘要
路径规划是汽车智能化技术研究的热点。算法延时及路径偏差大等问题是路径规划算法需要研究的重点问题。文章基于A^(*)算法,针对传统路径规划中存在的延时性问题,设计运行时间T_(pre);针对规划路径与车辆实际行驶路径偏差大的问题,提出优化的节点拓展方法,通过添加惩罚系数到估价方程,修改节点拓展计算方法,约束算法拓展节点的方向。设计硬件在环实验,验证改进算法的可行性。实验结果证实:算法耗时缩短4.23%;所规划的理论路径与车辆实际路径之间的平均偏差率从0.113%降至0.078%,优化后的A^(*)算法实时性和精确性均优于传统A^(*)算法。
Vehicle path planning is a hot spot of intelligent technology research.Algorithm delay and path deviation are the key problems for path planning algorithm.Based on A^(*) algorithm,aiming at the problem of delay in traditional path planning,this paper designed the running time T_(pre);aiming at the problem of large deviation between the planned path and the actual driving path of the vehicle,this paper proposed an optimized node expansion method,by adding penalty coefficient to the evaluation equation,modifying the node expansion calculation formula,and finally constraining the node expanding direction of the algorithm.We designed a virtual experiment to verify the feasibility of the improved algorithm.The experimental results show that the time-consuming of the algorithm is reduced by 4.23%;the average deviation rate between the planned theoretical path and the actual vehicle path is reduced from 0.113% to 0.078%.The real-time and accuracy of the optimized A^(*) algorithm are better than the traditional A^(*)algorithm.
作者
曾娟
刘为
张洪昌
ZENG Juan;LIU Wei;ZHANG Hong-chang(Hubei Key Laboratory of Advanced Technology for Automotive Components(Wuhan University of Technology),Wuhan 430070,China)
出处
《武汉理工大学学报》
CAS
北大核心
2019年第12期59-65,共7页
Journal of Wuhan University of Technology
基金
中央高校基本科研业务费专项资金(191007013).
