基于IT2SFNN的多关节机械手自适应反演控制

Adaptive Backstepping Control of the Multi-joint Manipulator Based on the IT2SFNN

考虑到机械手作业过程中的时变性和不确定性,针对多关节机械手模型模糊、干扰未知的问题,提出了一种基于区间二型顺序模糊神经网络(IT2SFNN)的自适应反演控制方法。在所提方案中,利用IT2SFNN对机械手模型的未知项进行估计,以降低建模误差对控制精度的影响;基于反演控制的方法对控制器进行设计,并针对神经网络的逼近误差和外部不确定干扰,设计了鲁棒补偿项;最后基于李雅普诺夫稳定性理论对系统稳定性进行证明,并通过对两关节机械手进行仿真分析,证实所提控制方案能够在保证控制精度的同时快速追踪到期望信号,且在受到干扰时具有较强的鲁棒性。

Considering the time variation and uncertainty of the manipulator,aiming at the problem of uncertain model and un⁃known disturbance of the multi-joint manipulator,a robust adaptive backstepping control method based on interval type-2 sequen⁃tial fuzzy neural network(IT2SFNN)is proposed.In the proposed scheme,the IT2SFNN is used to estimate the unknown terms of the manipulator model in order to reduce the influence of modeling error on the control accuracy.The controller is designed based on the backstepping control method,and a robust compensation term is designed for neural network approximation error and external uncertain disturbance.Finally,the stability of the system is proved based on Lyapunov stability theory,and through the simulation analysis of two-joint manipulator,so as to demonstrate that the proposed control scheme can quickly track the desired signal while ensuring the control accuracy,and shows strong robustness to disturbances.