考虑输出约束的机械臂自适应神经网络鲁棒控制

在工业机械臂系统的跟踪控制过程中,由于其结构和工作环境复杂,导致难以建立精确的系统模型,针对此问题提出了基于多层前馈神经网络的自适应鲁棒控制器.通过神经网络在线估计机械臂系统动力学模型,并在控制器中进行补偿,同时设计了一个在线更新的鲁棒项克服神经网络的重构误差;考虑机械臂实际系统的输出约束,采用障碍李雅普诺夫函数设计控制律并证明系统的稳定性从而使系统满足约束条件.仿真实验结果表明:在约束条件下所提出的控制器能够实现系统的一致最终有界稳定,且跟踪性能良好,并具有很好的抗干扰和自适应能力.

An Adaptive Neural Network Robust Control for Manipulator Considering Output Constraints

In the tracking control process of an industrial manipulator system,it is difficult to establish an accurate system model due to the complex structure and working environment of the manipulator system.An adaptive robust controller based on a multilayer feed-forward neural network is proposed to address this issue.The dynamic model of the manipulator system is estimated online through the neural network and compensated in the controller.At the same time,an online updated robust term is designed to overcome the reconstruction error of the neural network.Considering the output constraints of the actual manipulator system,the barrier Lyapunov function is used to design the control law and prove the stability of the system,which can make the system satisfy the constraint conditions.The simulation experiment results show that the proposed controller can achieve consistent and ultimately bounded stability of the system under the constraint conditions,and has good tracking performance,excellent anti-interference and adaptive capabilities.