基于人工免疫-改进粒子群优化算法的机械臂轨迹规划研究

Trajectory Planning of Manipulators Based on Artificial Immune-improved Particle Swarm Optimization Algorithm

焊接机器人运动轨迹复杂、控制精度要求高。提出了一种满足多目标约束的轨迹规划方法。针对机器人轨迹平滑性要求,以5次非均匀有理B样条(Non-Uniform Rational B-Splines,NURBS)曲线为基础,对笛卡儿空间路径点进行参数化表达;根据工业机器人路径约束及工况需求,选取时间、能耗、跃度3个运动学指标作为目标优化函数,采用人工免疫双态粒子群进行轨迹优化;为了平衡粒子“探索”与“利用”,增加双模态模型,引入人工免疫系统,提升了粒子多样性与后期收敛能力;根据Pareto解集得到满足约束的焊接机器人各关节最优轨迹,通过Matlab仿真证明了方法的有效性;最后,针对空间相贯曲线焊缝进行了焊接试验。结果显示,规划的轨迹符合实际工程需求。

The trajectory of the welding robot is complex and the control accuracy is high.A trajectory planning method is proposed to meet the multi-objective constraints.Aiming at the requirement of robot trajecto-ry smoothness,the Cartesian space waypoints are parameterized based on the quintic non-uniform rational B-splines(NURBS)curve.Based on the path constraints and operational requirements of industrial robots,three ki-nematic indicators of time,energy consumption,and jump are selected as the objective optimization functions,and artificial immune bimodal particle swarm is used for trajectory optimization.In order to balance the explora-tion and utilization of particles,a bimodal model is added,and an artificial immune system is introduced to in-crease the particle diversity and the later convergence ability.According to the Pareto solution set,the optimal trajectory of each joint of the welding robot satisfying the constraints is obtained,and the effectiveness of the method is proved by Matlab simulation.The results show that the planned trajectory meets the actual engineer-ing requirements.