摘要
针对传统算法无法适用多目标及动态环境的智能车路径规划问题,文中基于改进A*算法与势场蚁群算法进行了面向多目标的动态环境智能车路径规划算法研究。根据多目标的特征,采用改进A*算法识别完整的周边环境,并进行全局路径规划。对于实验场景中出现的局部变化或障碍物移动,将人工势场算法与蚁群算法相结合,获得了改进势场蚁群算法,以实现在原有全局路径规划基础上的局部修改。通过优化仿真得到了文中所提算法的最优参数值,并与蚁群算法进行对照测试。结果显示,所提算法相比对照组路径长度缩短了2.7%,具有良好的综合性能。
To solve the problem of the inability of traditional algorithm to fit the intelligent vehicle path planning in multi-objective and dynamic environment,this paper studies the intelligent vehicle path planning algorithm in multi-objective and dynamic environment based on improved A*algorithm and potential field ant colony algorithm.According to the characteristics of multi-objective,the improved A*algorithm is used to identify the complete surrounding environment and carry out global path planning.For the local changes or obstacles moving in the experimental scene,the artificial potential field algorithm and ant colony algorithm are combined to obtain the improved potential field ant colony algorithm to realize the local modification based on the original global path planning.The optimal parameters of the proposed algorithm are obtained through optimization simulation,and compared with ant colony algorithm.The results show that the path length of the proposed algorithm is shortened by 2.7%compared with the control group,thereby has good comprehensive performance.
作者
段焜
DUAN Kun(Jiangsu College of Safety Technology,Xuzhou 221011,Jiangsu Province,China)
出处
《信息技术》
2023年第6期66-70,共5页
Information Technology
基金
江苏省高等学校自然科学研究项目资助(17KJB-460003)。
关键词
多目标
动态环境路径规划
改进A*算法
改进势场蚁群算法
人工势场算法
multi-objective
dynamic environment path planning
improved A*algorithm
improved potential field ant colony algorithm
artificial potential field algorithm
