neural network robust adaptive control

The reference trajectory tracking is one of the most important issues in the field of tractor-trailer wheeled mobile robots control. In this paper, thetrajectory tracking control issues of a tractor-trailer wheeled mob ile robot has been significantly solved in the presence of structural uncertainties,non-hol o n o mic constraints and external disturbance. The proposed scheme is based on a design that the tractor-trailer’s state space representation is extracted from its dynamic and ki n e matic models and presented ina companion format first. In the following,by considering the state space representation of system, the control algorithm is presented includingtwo external and internal control loops. Toward this end, the control law has been developed in the inner loop via input-output feedback linearization in a nonlinear feedback formwh i ch is continuously eliminating the nonlinear dynamics of the system. Then,by using a comb ination of the output that is pr o duced in linearization steps with a terminal sliding mode control algorithm and sketching a neural robust ad aptive finite time controller in the outer loop, the accurate and fast performance of the closed loop system has been guar a nteed in the presence of uncertainties. The proposed control algorithmfinally guarantees the boundedness of closed-loop signals and accurate finite time convergence of tracking errors. At the end, the effectiveness of the proposed sch eme has been demo nstrated and shown through the extended Lyapunov theorem and simulated by MATLAB application.