一种关节型机器人快速收敛的粒子群优化算法

A Fast Convergent Particle Swarm Optimization Algorithm for Articulated Robots

快速收敛性的研究和应用对关节型机器人轨迹规划有着重要的意义和价值。以关节型机器人完成不同动作时轨迹的平滑度和所用的时间为主要优化目标,提出一种快速收敛的粒子群算法。将全局学习因子和局部学习因子联合起来,通过调整比例系数和全局学习因子的比重获得更快的收敛速度。实验结果表明:在迭代次数较少时,所提方法与改进的粒子群算法的应用效果接近;在迭代次数较多时,所提方法优于改进的粒子群算法。

Research and application of rapid convergence has important significance and value for the trajectory planning of articulated robots.Smoothness of the trajectory and the time used by the articulated robot for different actions were taken as main optimization objectives,a fast convergent particle swarm optimization algorithm was proposed.The global learning factor and the local learning factor were combined,and a faster convergence rate was obtained by adjusting the proportion coefficient and the proportion of global learning factor.The experimental results show that the application effect of the proposed method is close to that of the improved particle swarm optimization algorithm when the number of iterations is relatively small;the proposed method is better than the improved particle swarm optimization algorithm when the number of iterations is large.