基于遗传算法的机器人加工位姿优化研究

Research on Processing Posture Optimization of Robots Based on Genetic Algorithm

针对工业机器人的加工效果具有位姿依赖性这一特点,基于机器人性能评价指标和遗传算法提出一种优化机器人加工位姿的方法。建立机器人正运动学模型和静刚度模型并选取机器人运动性能评价指标和刚度性能评价指标;通过蒙特卡罗法得到机器人运动性能、刚度性能在关节空间和笛卡儿空间的分布图,通过分布图研究机器人各关节对机器人运动性能、刚度性能的影响;基于机器人运动性能评价指标和刚度性能评价指标设计适应度函数,使用遗传算法对机器人加工位姿进行优化,优化后机器人性能得到了提升。

To address the posture dependence of industrial robot in the machining process,a posture optimization methodology of robots based on performance evaluation indexes and genetic algorithm is presented.The conventional stiffness and kinematic models of the robots are established,and their performance indexes are presented.By Monte Carlo method,these two performance distribution maps in joint space and the Cartesian space are drawn to study the influence of each joint on these two performances.A fitness function is designed based on these two performance evaluation indexes,and genetic algorithm is used to optimaze processing posture of the robot,which brings about the improvement of the robot performance.