RBFNNBA Control of Electromechanical Servo System Based on the LuGre Model

To reduce the adverse effects of friction nonlinearity existing in electromechanical servo systems, an adaptive control method is proposed in this paper. First, the electromechanical servo system model is presented, followed by the introduction of the concept of virtual control quantity using the backstepping control method. The system's control law is designed by recursively selecting the Lyapunov function, and it is shown that system position tracking error is convergent. Because control law has unknown parameters, it is difficult to obtain accurate values for these parameters in practical engineering. Therefore, the universal approximation ability of a radial-basis-function (RBF) neural network is employed to approach the unknown parameters online. The system's adaptive law is designed, an RBF neural network backstepping adaptive controller based on state feedback is constructed, and the controller's stability is analyzed. Finally, a comparison is made between the simulation results of the proportional-integral-derivative (PID) controller, the RBF neural network (RBFNN) controller, and the RBF neural network backstepping adaptive (RBFNNBA) controller. Results show that performance of all three types of controller is gradually improved, which verifies the feasibility of the proposed control strategy.