Neural adaptive robust output feedback control of wheeled mobile robots with saturating actuators

Summary This paper addresses the output feedback tracking control problem of electrically driven wheeled mobile robots subjected to actuator constraints. The main drawback of previously proposed controllers is the actuator saturation problem, which degrades the transient performance of the closed-loop control system. In order to alleviate this problem, a saturated tracking controller has been proposed using the hyperbolic tangent function. A new nonlinear observer is introduced in order to leave out the velocity sensors in the robot system to decrease the cost and weight of the system for practical applications. A dynamic surface control strategy is effectively used to reduce the design complexity when considering actuator dynamics. In addition, neural network approximation capabilities and adaptive robust techniques are also adopted to improve the tracking performance in the presence of uncertain nonlinearities and unknown parameters. Semi-global stability of the closed-loop system is presented using direct Lyapunov method. Simulation results are provided to illustrate the effectiveness of the proposed control system for a differential drive mobile robot in practice. Copyright © 2014 John Wiley & Sons, Ltd.