基于二次优化的救援机器人路径规划
Path Planning of Rescue Robot Based on Quadratic Optimization
毛世勇 1李晓毅 1张小孟 2冯克涛 3朱刚 3王邠3
作者信息
- 1. 陆军工程大学 通信士官学校,重庆 400035
- 2. 31700 部队,辽宁 辽阳 111000
- 3. 31306 部队,四川 成都 610036
- 折叠
摘要
针对当前应急救援过程中救援机器人路径规划效率低的问题,采用改进的遗传算法和二次优化方法实现救援路径的优化.对传统遗传算法进行改进,在初始化过程中采用一种启发式可行点插入方法建立初始种群,在适应度函数中加入路径平滑评价函数,在保留策略中采取"抓大放小"的方法,使算法的全局收敛能力和收敛速度都有了较大的提高.在初始路线的基础上,由起点到终点采用逐段寻优的方法进行路径二次优化,进一步缩短规划路径的长度并减少拐点数量.仿真结果表明,该方法较次优方法,路径长度分别缩短 2.61%、2.21%、3.52%、1.22‰,平均收敛次数、搜索时间均有效优化,可有效地弥补传统遗传算法的缺陷,提高救援机器人的救援能力.
Abstract
Aiming at the low efficiency of rescue robot path planning in the current emergency rescue process,an improved genetic algorithm and quadratic optimization method is adopted to optimize the res-cue path.First,the traditional genetic algorithm is improved in the following way.In the initialization process,a heuristic feasible point insertion method is used to establish the initial population,a path smoot-hing evaluation function is added to the fitness function,and the method of"focusing on the big and relea-sing the small"is adopted in the retention strategy,which greatly improves the overall convergence ability and convergence speed of the algorithm.Secondly,on the basis of the initial route,the path is optimized section by section from the starting point to the end point to further reduce the length of the planned path and the number of inflection points.Finally,the method is simulated.The simulation results show that compared with the suboptimal method,the path length of this method is reduced by 2.61%,2.21%,3.52%and 1.22 ‰ respectively,and the average convergence times and search time are reduced,which can effectively solve the defects of traditional genetic algorithms and improve the rescue ability of rescue robots.
关键词
救援机器人/路径规划/遗传算法/二次优化Key words
rescue robot/path planning/genetic algorithm/quadratic optimization引用本文复制引用
出版年
2024
NETL
NSTL
维普
万方数据