基于改进粒子群算法的取件机械手轨迹综合优化设计

摘要 为优化取件机械手进入模具时的运动轨迹，提出了二次拉格朗日插值粒子群算法(QLIPSO)，此算法引入了二次拉格朗日插值局部搜索的方法，能够扩大搜索空间，避免局部收敛发生早熟，有效提高收敛精度.对比研究了典型粒子群算法(PSO)和几种改进型粒子群算法对取件机械手结构优化的设计效果.数值实验表明，QLIPSO算法具有最快的收敛速度，并且能够获得更好的优化结果.经该方法优化设计后，机械手在进入模具阶段的运动轨迹，与未优化前相比，直线度误差减少了98.06%，说明该方法能够有效优化取件机械手，获得更精确的运动轨迹.
Abstract：To optimize the motion curve for the pick-up manipulator when going into the mold， an novel method called quadratic Lagrange interpolation particle swarm optimization(QLIPSO)algorithm which can expand the search space， prevent premature convergence， and improve the convergence accuracy effectively is proposed. To verify the effectiveness of QLIPSO， four modified PSO algorithms are compared to synthesize the four-bar linkage of the manipulator. The numerical simulation results show that a better fitness value can be obtained with the QLIPSO algorithm. The linearity error of the curve of the manipulator， optimized by the proposed method， is decreased by 98.06% comparing with that non-optimized. It indicates that the proposed method can effectively improve the motion performance of the manipulator.