Three-Wheeled Mobile Robot Trajectory Tracking Control Using Nonlinear PID Controller Based Neural Network Combined With Backstepping Controller

Abstract

This paper presents a novel approach for the trajectory tracking control of a three wheeled mobile robot using neural network based nonlinear PID controller combined with the backstepping control techniques. First, the kinematic and dynamic model of the robot was derived using the Lagrangian method. Second, the neural network based nonlinear PID controller was trained and developed using the input and output data of a nonlinear PID controller(NPID). The nonlinear PID controller provided all the data used to train the proposed controller. The neural network has been trained using a Bayesian regularization algorithm. Finally, the stability of the designed controller is analyzed using the Lyapunov stability method and the performance of the proposed controller was compared with the Backstepping+PID controllers(BSC+PID) and BSC+NPID controller using MATLAB/Simulink. Extensive numerical simulations demonstrate that the BSC+NNNPID controller significantly enhances reference tracking accuracy, robustness and error minimizing capability compared to both BSC+PID and BSC+NPID when tested with different reference trajectory inputs.