基于改进粒子群算法的双工业机械臂避障路径规划

Research on obstacle avoidance algorithm for dual-industrial manipulators based on improved particle swarm optimization

为解决双机械臂工作过程中可能会发生碰撞的问题,提出包围盒算法的避障策略与改进粒子群算法的路径规划策略。先将机械臂与障碍物简化为线段与球体,建立线段到点的距离公式,随后采用基本粒子群算法(PSO)与改变惯性权重粒子群算法,依次规划出主、从机械臂的路径。改进粒子群算法内融入非线性函数e^(-x)与贝塔分布随机调整惯性权重。采用基本粒子群算法规划的路径长度分别为2.05 m与2.38 m;采用改进后的粒子群算法规划的路径长度分别为1.52 m与1.63 m。研究结果表明:改进的粒子群算法收敛精度更高、路径距离更短,满足安全性、路径的准确性要求。

To address the issue of possible collisions during the operation of dual robotic arms,this paper proposes an obstacle avoidance strategy using the bounding box algorithm and a path planning strategy using an improved particle swarm optimization algorithm.First,the manipulator and obstacles are simplified into line segments and spheres,and the distance formula between line segments and points is established.Then,the path of the master and slave manipulator is planned successively by using an elementary particle algorithm and particle swarm optimization(PSO) algorithm with changing inertia weight.The improved particle swarm optimization algorithm incorporates nonlinear function e~(-x) and beta distribution to adjust inertia weight randomly.After the path planning is completed,the path length of elementary particle swarm is 2.05 m and 2.38 m,respectively;the improved particle swarm optimization algorithm has path lengths of 1.52 m and 1.63 m,respectively.The research results indicate that the improved particle swarm optimization algorithm has higher convergence accuracy and shorter path distance and and meets the requirements of safety and path accuracy.