柔性机械臂振动与能量最小的多目标轨迹优化  

Vibration Suppression and Energy Minimization Trajectory Planning of Flexible Manipulator with Multi-Objective Optimization

作　　者：刘勇[1] 王腾 杜喆 LIU Yong;WANG Teng;DU Zhe(School of Mechanical Engineering Hefei University of Technology,Anhui Hefei 230009,China;Hefei University of Technology Xuancheng Campus,Anhui Xuancheng 242000,China)

机构地区：[1]合肥工业大学机械工程学院,安徽合肥230009 [2]合肥工业大学宣城校区,安徽宣城242000

出　　处：《机械设计与制造》2022年第8期297-300,304,共5页Machinery Design & Manufacture

基　　金：中央高校基本科研业务费专项资金项目(JZ2016HGBZ1042)。

摘　　要：为了同时减小柔性臂杆机械臂运动过程中的振动与能量消耗,提出了一种柔性机械臂多目标轨迹优化方法。首先,利用假设模态法和拉格朗日动力学方程对双柔性连杆机械臂进行动力学建模,进而利用MOPSO算法和传统近似方法求解同时满足振动最小和能量消耗最优的柔性参考轨迹;在此基础上,采用RBF神经网络进行逆动力学拟合,求解出连续平滑的关节驱动力矩,相应的关节轨迹满足任务需求,且进一步实现了振动抑制,有利于柔性机械臂的控制。最后,通过MATLAB仿真实验证明了所提算法的有效性。In order to optimize vibration during flexible manipulators’ motion and minimize energy cost simultaneously,a trajectory optimization method is proposed. 2-dof planar manipulator whose arms are both considered as flexible links is modeled with assumed modes method and Lagrangian approach dynamically. Reference trajectory that minimize vibration and energy consumption is calculated with MOPSO and traditional approximate approach. Inverse dynamic equation is fitted by RBF neural network to solve continuous and smooth joint input torque. The joint trajectories are able to meet the mission demands and vibration is further suppressed,which would be of benefit to control of flexible manipulators. The effectiveness is proved by MATLAB simulation.

关 键 词：柔性机械臂 柔性动力学 多目标轨迹优化 RBF神经网络 振动抑制 能量最小 

分 类 号：TH16[机械工程—机械制造及自动化] TP24[自动化与计算机技术—检测技术与自动化装置] TB32[自动化与计算机技术—控制科学与工程]