清洁机器人光伏电站复杂地形动态路径规划方法研究  

Research on Dynamic Path Planning Method for Cleaning Robots of Photovoltaic Power Plants in Complicated Topography

作　　者：王建军 王金鑫 张晨[1] 李翠明[1] WANG Jian-jun;WANG Jin-xin;ZHANG Chen;LI Cui-ming(School of Mechanical and Electrical Engineering,Lanzhou University of Technology,Lanzhou 730050,Gansu)

机构地区：[1]兰州理工大学机电工程学院,甘肃兰州730050

出　　处：《凿岩机械气动工具》2023年第4期10-18,27,共10页Rock Drilling Machinery & Pneumatic Tools

基　　金：甘肃省自然科学基金资助项目(18JR3RA139);甘肃省省级引导科技创新发展项目(2018ZX-13)。

摘　　要：针对移动机器人在光伏电站环境下的动态路径规划问题,基于光伏电站全局静态环境先验信息,提出一种改进A*蚁群系统算法的路径规划新方法。首先,引入改进估价函数的A*算法,通过调节权重系数,降低算法搜索工作量并解决蚁群系统(ACS)算法初次搜索的盲目性。其次,在ACS算法的基础上,设计群体返巢搜索策略以提高算法的收敛速度;运用自适应进化机制来优化解的生成,防止局部最优;当蚂蚁死锁时,提出分区惩罚因子以减少后续蚂蚁死锁的概率。最后在规划过程中,结合障碍物的大小和速度,提出五种防碰撞策略。仿真结果表明,该文算法和其他算法相比搜索到的路径质量更高,同时能指导清洁机器人有效躲避障碍物并规划出一条安全路径。Based on the prior information on the global static environment for a photovoltaic power plant,a method of the improved A*ant colony system algorithm was introduced for the dynamic path planning problem of mobile robots in the photovoltaic power plants environment.First of all,using the A*algorithm that improved the evaluation function,by adjusting the weight coefficient in the evaluation function of the A*algorithm,the algorithm search workload was reduced and the blindness of the initial search of the ant colony system algorithm was solved.Secondly,on the basis of the ant colony system algorithm,a return-to-nest search strategy was proposed to improve the convergence speed of the algorithm;the adaptive evolution mechanism was used to optimize the generation of the solution and prevent the local optimum;when the ant got lost,through the regional penalty to reduce the probability of subsequent ant getting lost.Finally,in the path planning,based on the size and speed of obstructions,5 anti-collision strategies were proposed.The simulation results show that the algorithm in this paper can find the better paths compared with other algorithms,enable the cleaning robots to successfully bypass obstructions,and plan a safe path.

关 键 词：光伏电站 清洁机器人 路径规划 蚁群系统算法 防碰撞策略 A*算法 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]