Study on Adaptive Sliding Mode Control of 2-DOF pan-tilt System Based on RBF Neural Network

Abstract

RBF neural network-based sliding mode control is a new nonlinear control strategy, which combines neural network and sliding mode controller, and can effectively solve some problems in nonlinear system control. In this paper, the proposed control algorithm is studied for 2-degree-of-freedom pan-tilt system, firstly, the kinematic and dynamics model of the system is derived, then the unknown system dynamics is modeled using RBF neural network into the sliding mode controller to realize the attitude control of 2-degree-of-freedom pan-tilt system. Compared with the traditional PID controller, the sliding mode controller based on RBF neural network exhibits better robustness and fast response performance in the nonlinear system. Since the RBF neural network has strong nonlinear approximation ability and self-adaptive capability, it can effectively overcome the model error and uncertainty of the nonlinear system. In the simulation experiments of this paper, we use MATLAB/Simulink tools, and verify the effectiveness of the proposed controller comparing with PID controller. The results show that the sliding mode controller based on RBF neural network proposed has better control performance in the attitude control of 2-degree-of-freedom pan-tilt system. In the future, the proposed control method can be applied to a real system to achieve better control effects.