Tracking control of a nonholonomic mobile robot by hybrid feedback and neural dynamics techniques

The mobile robot has been one of the challenging control topics due to its nonholonomic property and restricted mobility. The performance of a mobile robot highly depends on the tracking ability of the controller. This thesis presents the design of two path tracking controllers for a nonholonomic mobile robot. Neural dynamics technique is combined into both two-state and full-state conventional nonlinear feedback controllers. The proposed controllers are capable of generating real-time smooth forward and angular velocities, which drive a mobile robot to track the desired trajectories. They successfully resolve the speed jump problem which has plagued in many previous tracking controllers. Moreover, the proposed controllers can track not only the postures of a moving reference mobile robot, but also the discrete trajectories that do not meet the nonholonomic constraint requirements. Lyapunov stability theory is used to establish the stability of the proposed control system and the convergence of tracking errors toward zero. Various simulations and comparative studies have demonstrated the effectiveness of the proposed controllers. Both controllers can be integrated with a path planner to guide the navigation of an autonomous mobile robot. A neural dynamics based path planner generates a collision-free trajectory for a point robot in an unknown environment. The nonholonomic constraints of the mobile are considered in the proposed tracking controllers. The mobile robot is capable of avoiding the deadlocks, and generating a continuous path with smooth velocities. Simulation results of the synthesis for the path planner and tracking controller are also presented in this thesis.