基于遗传算法的机器人自定位、路径规划研究  

Research on Robot Self-localization and Path Planning Based on Genetic Algorithm

作　　者：韩菊 HAN Ju(Taiyuan University,Taiyuan 030024,Shanxi)

机构地区：[1]太原学院,山西太原030024

出　　处：《电脑与电信》2022年第12期77-81,共5页Computer & Telecommunication

摘　　要：利用栅格图法,构建移动机器人位置定位、路径规划的环境模型,在此基础上依托实值编码方式,围绕初始位姿、路径长度、拐点数量等评估指标进行初始化种群编码,将被选中的栅格连成一条完整路径,并根据声呐回传位置信息、路径信息相似度,从种群中选择与真实位姿、路径相近的算子个体,并进行算子的自适应选择、交叉和变异计算,经过多代遗传迭代进化后,得到适应度较高的新的个体,以及机器人的最优定位位置、最短路径长度和拐点数量。仿真实验及结果得出:改进遗传算法对机器人位置定位、路径规划和拐点数量计算,算法迭代的次数更少、收敛度更优,有效提高了机器人自定位及路径规划平滑度、算法收敛速度。By using the raster graph method,the environment model of mobile robot position positioning and path planning is constructed.Based on this,the population coding is initialized around the evaluation indexes such as initial pose,path length and number of inflection points.The selected grid is connected to a complete path,and the location information and path information similarity are transmitted back according to sonar.The operator individuals close to the real pose and path are selected from the population,and the operators are self-adaptive selection,crossover and mutation calculation.After multi-generation genetic iterative evolution,the new individuals with high fitness are obtained,as well as the optimal positioning position,the shortest path length and the number of inflection points of the robot.Simulation experiments and results show that the improved genetic algorithm for robot position positioning,path planning and number of inflection points calculation,algorithm iteration times is less,the convergence is better,effectively improve the robot self-positioning and path planning smoothness,algorithm convergence speed.

关 键 词：移动机器人 全局定位 路径规划 遗传算法 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]