基于双粒子群算法的矿井搜救机器人路径规划

针对在复杂地形中标准的粒子群算法用于矿井搜救机器人路径规划存在迭代速度慢和求解精度低的问题,提出了一种基于双粒子群算法的矿井搜救机器人路径规划方法。首先将障碍物膨胀化处理为规则化多边形,以此建立环境模型,再以改进双粒子群算法作为路径寻优算法,当传感器检测到搜救机器人正前方一定距离内有障碍物时,开始运行双改进粒子群算法:改进学习因子的粒子群算法(CPSO)粒子步长大,适用于相对开阔地带寻找路径,而添加动态速度权重的粒子群算法(PPSO)粒子步长小,擅长在障碍物形状复杂多变地带寻找路径;然后评估2种粒子群算法得到的路径是否符合避障条件,若均符合避障条件,则选取最短路径作为最终路径;最后得到矿井搜救机器人在整个路况模型中的最优行驶路径。仿真结果表明,通过改进学习因子和添加动态速度权重提高了粒子群算法的收敛速度,降低了最优解波动幅度,改进的双粒子群算法能够与路径规划模型有效结合,在复杂路段能够寻找到最优路径,提高了路径规划成功率,缩短了路径长度。

Path planning of mine search and rescue robot based on two-particle swarm optimization algorithm

In view of problems of slow iterative speed and low solution accuracy of standard particle swarm optimization algorithm used in the path planning of mine search and rescue robot in complex terrain,a path planning method for mine search and rescue robot based on two-particle swarm optimization algorithm was proposed.Firstly,the obstacles are expanded into regular polygons to build an environment model,and then the improved two-particle swarm optimization algorithm is used as the path optimization algorithm.When the sensor detects obstacles within a certain distance in front of the search and rescue robot,it starts to run the improved two-particle swarm optimization algorithm:particle swarm optimization algorithm with improved learning factor(CPSO)grows in steps,which is suitable for finding paths in relatively open areas,while particle swarm optimization algorithm with dynamic velocity weight(PPSO)has small particle steps,which makes it good at finding paths in complex and variable areas of obstacle shapes.Then the algorithm evaluates the paths obtained by the two particle swarm optimization algorithms whether meet the obstacle avoidance requirements or not.If both meet the obstacle avoidance requirements,the shortest path is selected as the final path.Finally,the optimal driving path of the mine search and rescue robot in the whole road condition model is obtained.The simulation results show that the convergence speed of particle swarm optimization algorithm is improved by improving the learning factor and adding the dynamic velocity weight,and the optimal solution fluctuation range is reduced;the improved two-particle swarm optimization algorithm can be effectively combined with the path planning model,and the optimal path can be found in the complex road section,which improves the success rate of path planning and shortens the path length.