Tracking Control of Wheeled Mobile Robots Using Fuzzy CMAC Neural Networks

Tracking Control of Wheeled Mobile Robots Using Fuzzy CMAC Neural Networks

This paper investigates the trajectory tracking control problem of wheeled mobile robots. First, the analytic Bspline function is used to generate a smooth feasible trajectory between the initial and the desired configurations so that the motion path can pass through the desired intermediate points to satisfy the kinematic constraints and curvature restrictions. Given the desired B-spline trajectory, the corresponding reduced dynamics can be used to design the control law for the privileged coordinates. Although the privileged coordinates can be driven to the desired values, the postured coordinates may significantly deviate from the reference values if the initial conditions are not appropriately provided, or there are disturbances during the motion. To solve this problem, the fuzzy control results are presented from the practical experience of driving to steer the postured coordinates in a kinematic level. To enhance the robot tracking performance and assure the error convergence, the robust adaptation laws for the FCMAC and compensated controller are derived from the stability analysis. Finally, an illustrated example shows the performance of the proposed trajectory tracking robot control scheme.