中文摘要：

提出一种新型的多目标优化遗传算法,该算法采用两种精英机制,加快了收敛速度,避免了在一般多目标遗传算法中难以处理的适应值分配过程,减小了计算资源的消耗。把所提算法应用于带有N个关节的冗余机械手运动学逆解问题,与传统的机械手逆解方法相比,所提算法不仅能够使得终端执行器精确到达期望位置,而且同时优化了机械手关节转动角度、柔顺性、安全性三个目标。仿真结果表明了所提算法的有效性。

英文摘要：

A new multi-objective optimization genetic algorithm was proposed.It employed two elitism mechanisms to motivate convergence and avoided the procedure of fitness assignment which was a difficult design issue in general genetic multi-objective optimization.The proposed algorithm was applied to the inverse kinematics problem of the redundant manipulator.Compared to traditional methods,it not only made the end-effector arrive the desired position precisely,but also optimized three objectives which were turning angles of joints,flexibility and safety of the manipulator.Simulation results indicate the efficiency of the proposed algorithm.