Adaptive Fuzzy Predictive Hybrid Position/Velocity Control of Nonholonomic Wheeled Mobile Robots

This paper presents Nonlinear Model-based Predictive Control (NMPC) of Wheeled Mobile Robots (WMRs) based on a discrete-time fuzzy model to approximate the dynamics of the robot and the actuator. The parameters of the fuzzy model are adjusted on-line by using gradient descent algorithm and recursive least square estimation method in order to cope with uncertainties in the system. Moreover, by tuning the weights in the cost function of the NMPC, better tracking error of the WMR can be obtained. The simulation results show that the proposed method can effectively control a type (2,0) WMR with a good performance

Keywords:

Mobile robots , Trajectory tracking , Nonlinear model predictive control , Fuzzy modeling , Adaptive control