论文部分内容阅读
针对具有建模误差和不确定干扰的机械臂轨迹跟踪控制,提出了一种改进的自适应神经滑模变结构控制方法。该方法在传统神经滑模变结构控制的基础上,设计了一种对神经网络学习误差的自适应补偿控制,以补偿神经网络在初始阶段可能出现的误差,增强了系统的稳定性,实现了滑模变结构控制的切换控制增益对建模误差和不确定干扰的自动跟踪,削弱了抖振。利用李亚普诺夫定理证明了控制系统的稳定性,并利用仿真实验比较和分析了所提出方法与传统神经滑模变结构控制的运行结果。仿真结果表明该方法有效地削弱了滑模变结构控制的抖振,并具有更好的控制性能。
Aimed at the robot trajectory tracking control with modeling errors and uncertainties, an improved adaptive neural-sliding variable structure control method is proposed. Based on the traditional sliding mode control of neural networks, this method designs an adaptive compensation control algorithm for neural network learning error to compensate for the possible error in the initial stage of neural network and enhance the stability of the system. Switched-mode sliding mode control The gain of the control mode automatically tracks the modeling error and the uncertain interference, which weakens the chattering. The stability of the control system is proved by using Lyapunov theorem. The simulation results show that the proposed method and the traditional sliding mode control with variable structure are compared and analyzed. Simulation results show that this method effectively weakens the chattering of sliding mode variable structure control and has better control performance.