Position and Orientation Control of a Mobile Robot Using Neural Networks

In this paper, an adaptive neuro-control system with two levels is proposed for the motion control of a nonholonomic mobile robot. In the first level, a PD controller is designed to generate linear and angular velocities, necessary to track a reference trajectory. The proposed strategy is based on changing the robot control variables. Using this model, the nonholonomic constraints disappear and shows how the direct adaptive control theory can used to design robot controllers. In the second level, a neural network converts the desired velocities, provided by the first level, into a torque control. By introducing appropriate Lyapunov functions asymptotic stability of state variables and stability of system is guaranteed. The tracking performance of neural controller under disturbances is compared with PD controller. Sinusoidal trajectory and lemniscate trajectories are considered for this comparison.