Sliding-Mode-Disturbance-Observer-Based Robust Tracking Control for Omnidirectional Mobile Robots with Kinematic and Dynamic Uncertainties

Abstract

In this article, a robust tracking control method is developed for omnidirectional mobile robots (OMRs) with uncertainties in the kinematics and dynamics. The kinematic and dynamic uncertainties that significantly degrade the OMR tracking control performance should be simultaneously compensated, which has not been achieved by the existing OMR tracking control methods. Therefore, the OMR dynamics containing the actuator dynamics and dynamic uncertainties are obtained by including the friction present in the OMR and are identified using a system identification method for the actual OMR system. The kinematic uncertainties present in the OMR kinematics are then derived unlike the existing studies that do not consider them. A sliding mode disturbance observer is proposed to estimate these kinematic uncertainties. A robust backstepping-like feedback linearization tracking controller using estimates of both the kinematic and dynamic uncertainties is also proposed to compensate for these uncertainties. The proposed method is validated through a stability analysis and simulation and experimental results using an actually implemented OMR system.