针对煤矿救援机器人在井下巷道中进行搜救时,存在搜索效率低且易发生剐蹭的问题,选择煤矿救援机器人为研究对象,提出了一种基于智能仿生算法的路径规划解决方案,通过仿真实验开展井下路径规划研究,分析路径规划算法在井下应用面临的问题;简述路径规划算法的分类和使用场景,比对分析算法的优缺点;分别选择采用PSO粒子群算法、GWO灰狼算法和GA遗传算法作为煤矿救援机器人路径规划的基本算法,并对算法原理进行简要分析。为验证三种算法路径规划的优劣,进行MATLAB仿真实验,构建三张不同障碍物地图来模仿机器人在复杂路径下的行驶路线。结果表明:PSO算法最短路径相较GWO算法和GA算法分别下降了7.05%和2.85%。粒子群算法在稳定性、路径规划精度方面更高,满足井下救援需求。

In view of the existing problem of low search efficiency and easy occure scratching of coal mine rescue robots during search and rescue in downhole roadways, choosing coal mine rescue robots as the research object, a path planning solution scheme based on intelligent biomimetic algorithm is proposed. Through simulation experiments, downhole path planning research is carried out to analyze the problems faced by path planning algorithm in downhole applications; Describe the classification and usage scenarios of path planning algorithm, compare and analyze the advantages and disadvantages of the algorithm; Choose and adopt PSO particle swarm optimization algorithm, GWO grey wolf algorithm, and GA genetic algorithm, respectively, as the basic algorithms for path planning of coal mine rescue robots, and briefly analyze the algorithm principles; To verify the advantages and disadvantages of three algorithms for path planning, MATLAB simulation experiments are conducted to construct three different obstacle maps to simulate the driving routes of robots under complex paths. The results show that the PSO algorithm reduces the shortest path by 7.05% and 2.85% compared to the GWO algorithm and GA algorithm, respectively, the particle swarm optimization algorithm is higher in aspects of stability and path planning accuracy, meeting the needs of downhole rescue.