基于GA的时间最优机械臂轨迹规划算法  

Time-Optimal Trajectory Planning Algorithm Based on GA for Manipulator

作　　者：居鹤华[1] 付荣[1] 

机构地区：[1]北京工业大学电子信息与控制工程学院,北京100124

出　　处：《控制工程》2012年第3期472-477,共6页Control Engineering of China

基　　金：国家863项目(2008AA0085);国家自然科学基金(60975065)

摘　　要：由于多项式插值轨迹规划具有阶次高、没有凸包性质等特点,传统优化方法难以应用的特点,根据机械臂运动学约束,提出了关节空间基于遗传算法(GA)的3-5-3多项式插值轨迹规划算法。利用运动学约束,以最优时间为目标,针对关节型机器人在静态环境下的点到点的轨迹规划问题,利用GA算法解算多项式插值的时间。通过与基于PSO的3-5-3多项式机械臂轨迹规划运动位置、速度、加速度曲线对比,证明该方法在运行时间和运行平稳度上都有突出优点。Since polynomial is high order and has not share the characteristic of convex hull, it is hard to be optimized by traditional optimization methods. According to the velocity limitation of manipulator, a Time-optimal 3-5-3 polynomial interpolation trajectory plan- ning algorithm based on Genetic Algorithm （GA） is proposed. To plan the point-to-point trajectories for manipulator working in a static environment for time-optimal, GA is used to get the time of polynomial interpolation. The simulation results show that the algorithm has good performance on time-optimal and stability compared with 3-5-3 polynomial interpolation trajectory planning algorithm based on PSO.

关 键 词：机械臂 轨迹规划 时间最优 遗传算法 多项式插值 

分 类 号：TP27[自动化与计算机技术—检测技术与自动化装置]