Set-Membership Filtering Based Model Predictive Control for Trajectory Tracking of Automated Guided Vehicles

This article considers the trajectory tracking problem of automated guided vehicles with unknown-but-bounded noises and proposes a set-membership filtering (SMF) based model predictive control strategy. A 2-DOF vehicle model is first established with considering the unknown-but-bounded process and measurement noises. Then, a trajectory tracking control scheme is proposed which consists of two parts, that is, a set-membership observer and a model predictive controller. The proposed set-membership observer is able to deal with the unknown-but-bounded noises and obtain the state estimation ellipsoid for the considered automated guided vehicle. Further, the designed model predictive controller utilizes the estimated ellipsoid to compute the control sequence. Finally, simulation results demonstrate the effectiveness and superiority of the proposed method.