摘要
针对A*算法用于机器人驾驶汽车路径规划时冗余点和效率问题,提出一种变网格A*算法。先改进评价函数的计算方式和代价计算方式,从而降低搜索时间。再针对不同情况自适应调整不同的搜索邻域和不同的评价函数权重,保证搜索效率的同时兼顾路径平滑度。综合考虑汽车的行驶安全性和规划路径的平滑度,对障碍物进行膨胀化处理,保证了汽车转弯过程中的安全距离。同时对路径进行提取关键节点处理和可调节圆弧优化处理,针对不同汽车采用不同的转弯半径,保证了汽车行驶的平滑度和可操作性。最后为了验证变网格A*算法的性能,对上述算法进行仿真对比与实验验证。实验结果表明,路径平滑度、运算效率均有明显提高。
To solve the redundancy and efficiency problems of A* algorithm in path planning of robot-driven vehicles,a variable grid A* algorithm was proposed in the paper.Firstly,the calculation method of evaluation function and cost calculation method were improved to reduce the search time.Secondly,different search neighborhoods and different evaluation function weights were adaptively adjusted according to different situations to ensure both search efficiency and path smoothness.Then,the safety of the vehicle and the smoothness of the planned path were considered comprehensively,and the obstacle was inflated to ensure the safe distance of the vehicle in the process of turning.At the same time,the key nodes of the path were extracted and the adjustable arc was optimized.Different turning radii were adopted for different vehicles to ensure the smoothness and maneuverability of the vehicle.Finally,in order to verify the performance of the variable grid A* algorithm,the simulation comparison and experimental verification of the algorithm were carried out.The experimental results show that the smoothness of the path and the calculation efficiency are improved obviously.
作者
任荣伦
卫沅
周志刚
李浩然
REN Rong-lun;WEI Yuan;ZHOU Zhi-gang;LI Hao-ran(Henan University of Science and Technology,Luoyang Henan 471000,China)
出处
《计算机仿真》
北大核心
2023年第9期443-448,共6页
Computer Simulation
基金
国家自然科学基金(51805149)
河南省重点研发与推广专项(202102210317)。
关键词
变网格
评价函数权重
可调节圆弧优化
Change grid
Weight of evaluation function
Adjustable arc optimization
