基于神经网络状态观测器的反演控制在双关节机械手控制中的应用

Back-stepping Position Controller for Two-joint Manipulator Based on Neural Network State Observer

提出了一种基于改进的BP神经网络的自适应状态观测器,该类观测器无需系统的精确模型即可得到收敛于真实状态的状态观测值。利用Lyapunov直接法分析了基于状态输出误差的状态观测器的稳定性。然后,将状态观测器与反演控制器分开设计,以实现观测器得到的速度估计值代替实际速度,避免了实际应用中对速度信号的测量。最后通过对二关节机械手系统的仿真与比较,说明该控制方法的有效性。

An adaptive state observer based on modified BP neural network was developed, which can converge the observed state to the truth state without exact knowledge of the nonlinear system. The stability of the state observer was analyzed by Lyapunov direct method. Then the state observer and back-stepping controller were designed respectively, so the estimated velocity of observer substitu- ted the actual one, and avoided measurement of velocity signal in actual application. Finally, the effectiveness of control method is veri- fied by simulation and comparison of the two-joint manipulator system.