A Novel Adaptive Sliding Mode Control of Robot Manipulator Based on RBF Neural Network and Exponential Convergence Observer

This paper focuses on a novel adaptive sliding mode control (NASMC) of robot manipulator based on RBF (radial basis function) neural network and observer. A novel adaptive sliding mode control can achieve high performance tracking control by designing three adaptive parameters. Different from other existing adaptive control methods, an exponential convergence observer is designed to solve the parameter uncertainty, and the unknown nonlinear friction can be obtained by the online estimation of RBF neural network. Then the observer value and the RBF neural network estimation value are transferred to the controller, and the equivalent compensation is introduced to realize the stable control of the system. By utilizing Lyapunov stability theory, it is proved that the system can realize adaptive control under the designed controller. The effectiveness of the control method is verified by simulation. The amount of operation can be reduced through the NASMC method, and the value of root mean squared error is 0.00031795, which is closer to 0. Compared with adaptive sliding mode control (ASMC) and RBF neural network adaptive control (RBFAC), the robot manipulator system has better tracking effect.