Modified Model Predictive Control of Wheeled Mobile Robot Using Adaptive Reference Point and Prediction Horizon

Under certain circumstances, the curvature of the reference path for a mobile robot can change sharply, leading to reduced control accuracy. To improve the trajectory tracking capability of a mobile robot, this paper proposes a modified model predictive control (MMPC) method with an adaptive reference point and prediction horizon. The design begins with constructing the single-track kinematic model of the wheeled mobile robot (WMR). Then, an adaptive law is established to enable the reference point to vary with the current path curvature. Specifically, the prediction horizon is flexibly modified according to the distance between the WMR and the reference point. The verification experiments demonstrate the effectiveness and robustness of the proposed MMPC method compared to conventional methods