基于GA-PSO的矿井巡检机器人路径规划

Path planning of mine inspection robot based on GA-PSO

在复杂矿井地形中,标准的粒子群算法(Particle Swarm Optimization,PSO)和遗传算法(Genetic Algorithm,GA)在巡检机器人路径规划方面分别存在局部寻优能力贫乏、全局寻优能力欠缺的问题,针对这一问题,提出了一种基于GA-PSO算法的矿井巡检机器人路径规划新算法。该算法先将障碍物在栅格地图上表示为规则图形,完成对环境模型的创建;再引入随机惯性权重模型优化标准粒子群算法,通过在粒子群算法中加入遗传算法的选择、交叉和变异操作算子,将得到遗传信息的精英粒子群分割出来产生下一代粒子,完成算法的融合,以此作为路径寻优算法使机器人完成起点到目标点的路径搜索。仿真结果表明,新算法相较于原算法成功率显著提高且路径长度明显缩短,即新算法的综合性能优于两种算法单独使用的效果,使得全局路径规划结果所得的路线更为平滑。

To address the problem that the standard particle swarm optimization(PSO)and genetic algorithm(GA)have poor local optimisation capability and lack of global optimisation capability respectively for path planning of inspection robots in complex mine terrains,a GA-PSO based path planning method is proposed.GA-PSO based path planning method for mine inspection robots.Firstly,the obstacles are represented as regular graphs on the raster map,which completes the creation of the environment model;then the standard particle swarm algorithm is optimised by introducing the random inertia weight model,and by adding the selection,crossover and variation operators of the genetic algorithm to the particle swarm algorithm,the elite particle swarm with genetic information is partitioned to generate the next generation of particles,which completes the fusion of the algorithms;this is used as the path finding algorithm to enable the robot to complete the path search from the starting point to the target point.The simulation results show that the success rate of the hybrid algorithm is significantly improved compared with the original algorithm and the path length is significantly shortened,i.e.the comprehensive performance of the hybrid path planning algorithm proposed in this paper is better than that of the two algorithms alone,resulting in a smoother route for the global path planning results.