基于PSO算法的水处理罐搬运机器人时间最优轨迹规划

PSO-based Trajectory Planning for Water Treatment Tank Handling Manipulator

针对传统水处理罐搬运机器人轨迹规划效率低下、运行不稳定等问题,提出了一种引入动态调整学习因子和惯性因子的粒子群算法,对分段多项式插值轨迹进行优化。该方法简化了传统粒子群算法中自适应函数的构建方式,改善了传统粒子群算法早期容易陷入局部最优解、后期收敛速度慢等问题。通过仿真对比实验获得机械手各关节的位置、速度和加速度轨迹,证明该方法能够提高机器人运行效率,同时保证整体运行的稳定性,有效实现了机器人的轨迹优化。

To tackle the issues of low trajectory planning efficiency and unstable operation for traditional water treatment tank handling robots,a particle swarm(PSO)algorithm that introduced dynamic adjustment learning factor and inertia factor was proposed.This method effectively combines the piecewise polynomial function with the modified PSO algorithm,which simplifies the construction of the adaptive function.The simulation experiment shows that the developed method can realize the time-optimized robot trajectory,and ensure the stability of the overall operation while improving the work efficiency.