基于PD型迭代学习的机械臂轨迹跟踪控制

Trajectory Tracking Control of Manipulator Based on PD Type Iterative Learning

为实现机械臂对期望轨迹的精确跟踪或完成特定的位姿变换,结合迭代学习算法与滑模控制算法的特点,提出一种迭代控制与滑模控制相融合的控制策略。采用拉格朗日法对机械臂进行动力学建模,设计PD型迭代滑模算法,并使用Lyapunov理论对控制算法进行收敛性分析。在MATLAB/Simulink软件中以一个二自由度机械臂为例,在不同初始条件和扰动的情况使用该控制策略进行机械臂轨迹跟踪仿真实验。结果表明,该控制策略能在机械臂迭代初值不一致的情况下实现对期望轨迹的精确跟踪,PD型迭代滑模控制与PD型迭代学习控制相比具有误差收敛速度快、跟踪精度高和鲁棒性强等优点。

In order to accurately track the desired trajectory or complete the specific pose transformation of the manipulator,combined with the characteristics of iterative learning algorithm and sliding mode control algorithm,a control strategy combining iterative control and sliding mode control is proposed.The dynamics of the manipulator is modeled by Lagrange method,the PD type iterative sliding mode algorithm is designed,and the convergence of the control algorithm is analyzed by Lyapunov theory.In MATLAB/Simulink software,taking a two degree of freedom manipulator as an example,the control strategy is used to simulate the trajectory tracking of the manipulator under different initial conditions and disturbances.The results show that the control strategy can accurately track the desired trajectory when the initial iterative values of the manipulator are inconsistent.Compared with PD iterative learning control,PD iterative sliding mode control has the advantages of fast error convergence,high tracking accuracy and strong robustness.