基于改进蚁群和DWA算法的机器人动态路径规划  

Robot dynamic path planning based on improved ant colony and DWA algorithm

作　　者：魏立新[1,2] 张钰锟 孙浩[1,2] 侯仕杰 WEI Li-xin;ZHANG Yu-kun;SUN Hao;HOU Shi-jie(Engineering Research Center of the Ministry of Education for Intelligent Control System and Intelligent Equipment,Yanshan University,Qinhuangdao 066004,China;Key Laboratory of Industrial Computer Control Engineering of Hebei Province,Yanshan University,Qinhuangdao 066004,China;Anhui Nari Jiyuan Electric Power System Tech Co.Ltd.,Hefei 230000,China)

机构地区：[1]燕山大学智能控制系统与智能装备教育部工程研究中心,河北秦皇岛066004 [2]燕山大学工业计算机控制工程河北省重点实验室,河北秦皇岛066004 [3]安徽南瑞继远电网技术有限公司,合肥230000

出　　处：《控制与决策》2022年第9期2211-2216,共6页Control and Decision

基　　金：国家重点研发计划项目(2018YFB1702300);国家自然科学基金项目(62003296);河北省青年基金项目(E2018203162)。

摘　　要：路径规划技术是移动机器人研究领域中的一个重要分支,使得机器人能够在多障碍物环境中安全快速地找到一条相对最优路径.针对全局路径规划时蚁群算法盲目性搜索、易陷入局部最优、收敛速度慢以及局部路径规划时DWA算法难以有效地规避动态障碍物等问题,提出一种改进蚁群算法与DWA算法的融合算法.首先,采用GRRT-Connect算法不等分配初始信息素,解决陷阱地图中局部最优问题;然后,增加蚁群接力搜索方法以解决蚂蚁禁忌表自死锁问题,并利用切片取优方法优化最优路径选择机制得到全局最优路径;接着,以最优路径关键点为子目标点运行DWA算法,提出自适应调节速度方法进行最优行驶;最后,提出预计算方法规避动态障碍物达到局部规划效果.仿真结果表明,与现有文献结果相比,融合算法最优路径长度缩短了10.28%,收敛速度加快了6.55%,验证了所提出算法的有效性和优越性.Path planning technology is an important branch in the field of mobile robot research,which enables the robot to find a relatively optimal path safely and quickly in the multi-obstacles environment.Aiming at the blind search of the ant colony algorithm,easy to fall into local optimization and slow convergence speed in global path planning,and the problem that the dynamic window approad(DWA)is difficult to effectively avoid dynamic obstacles in local path planning,a fusion algorithm of the improved ant colony algorithm and the DWA is proposed.Firstly,the GRRT-Connect algorithm is proposed to allocate initial pheromones unequally to solve the local optimization problem in trap maps.Secondly,the ant colony relay search method is added to solve the self deadlock problem of an ant tabu list,and the slice optimization method is used to optimize the optimal path selection mechanism to obtain the global optimal path.Then,the DWA is run with the key points of the optimal path as the sub-target points,and an adaptive speed adjustment method is proposed for optimal driving.Finally,a pre-calculation method is proposed to avoid dynamic obstacles and achieve the effect of local planning.The simulation results show that compared with the results in the existing literature,the optimal path length of the fusion algorithm is shortened by 10.28%and the convergence speed is accelerated by 6.55%.

关 键 词：路径规划 移动机器人 蚁群算法 蚁群接力 DWA算法 动态障碍物 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]