Abstract: To improve the poor convergence quality of Clonal Selection Algorithm (CSA) in solving complex optimization problems,an improved Clonal Selection Algorithm based on Hybrid Mutation (HMCSA) was proposed,which was used to solve the inverse kinematics problem of redundant manipulator.A two-level hybrid cooperative search mode of hybrid differential mutation and clonal selection evolution was adopted to balance the global exploration and local exploitation of the algorithm effectively.Thus,the premature convergence problem of the basic clonal selection algorithm was well overcome.Through the dynamic real-time sharing of successful antibody search experience,the convergence speed of the algorithm was accelerated,and the convergence accuracy of the algorithm was also improved.In addition,compared with the comparison algorithms,the HMCSA had the advantage that fewer parameters needed to be set,which was convenient for the application and promotion of the algorithm.The effectiveness of HMCSA was verified by the classical benchmark function optimization problems.In the optimization of inverse kinematics solution of redundant manipulator,HMCSA also achieved good optimization results,which provided a new idea for solving inverse kinematics problem of robot.

Key words: redundant manipulator, inverse kinematics, clonal selection algorithm, hybrid mutation

摘要： 为改善克隆选择算法(CSA)在解决复杂优化问题时收敛质量不高的不足,提出一种基于混合变异的改进克隆选择算法(HMCSA),并将该算法用于解决冗余机械臂的逆运动学问题。改进算法采用了混合差分变异与克隆选择进化的两级混合协同搜索模式,有效平衡了算法的全局探索与局部开发,从而较好地克服了基本克隆选择算法容易陷入局部极值而早熟收敛的现象。通过抗体成功搜索经验的动态实时共享,加速了算法的收敛速度以及提升了算法的收敛精度。此外,与对比算法相比,HMCSA算法具有需要设置参数更少的优点,便于算法的应用推广。通过经典的基准测试函数优化问题验证了HMCSA算法的有效性;在冗余机械臂运动学逆解的优化求解中,HMCSA算法同样获得了较好的优化效果,为解决机器人的逆运动学问题提供了新思路。

关键词: 冗余机械臂, 逆运动学, 克隆选择算法, 混合变异