基于精英策略粒子群算法机械臂逆运动求解  

Solution of Mechanical Arm Inverse Kinematics Based on ETMOPSO

作　　者：杨欢[1,2] 李培林[1] 曹钧[2] 孙野[2] 陈忠凯[2] 

机构地区：[1]空军工程大学,陕西西安710051 [2]西北核技术研究所,陕西西安710024

出　　处：《测控技术》2016年第10期59-61,65,共4页Measurement & Control Technology

摘　　要：多关节机械臂空间姿态控制研究中,逆运动分析是控制的基础,由于多关节机械臂逆运动求解问题较为复杂,对其求解结果优劣性的判别比较困难。为了解决传统关节变量算法在求解过程中收敛精度不高、局部最优解过多的问题,利用精英策略搜索粒子种群中的优质粒子,提高了算法整体适应度值,避免求解陷入局部最优解。仿真实验结果表明,基于精英策略多目标粒子群算法(ETMOPSO)提高了机械臂空间姿态的求解精度,为多关节机械臂空间姿态控制提供了一种研究思路。Inverse kinematics analysis is the basis of the multi-joint mechanical arm space attitude control study. The discrimination of solving results superiority-inferiority is difficult, because the multi-joint manipulator inverse kinematics is more complex. In order to solve the problems that low convergence accuracy and too many the local optimal solutions of the traditional joint variable algorithm in the process of solving, the elite tactic is used to search the high quality particle in the particle population. This algorithm improves the overall fitness value, avoids solving falling into the local optimal solution. The simulation results show that the ETMOPSO（ elitist tactics multi-objective particle swarm optimitation） improves the attitude precision of mechanical arm space position, at the same time it provides a research way for multiple joint robotic arm space position control.

关 键 词：多关节机械臂 精英策略 粒子群算法(PSO) 仿真实验 

分 类 号：TH113.2[机械工程—机械设计及理论]