Slip ratio estimation and control of wheeled mobile robot on different terrains

This paper presents a model-based algorithm for estimating the longitudinal velocity and online slip ratio control of wheeled mobile robots (WMR). The adaptive unscented Kalman filter (AUKF) is employed to estimate the vehicle longitudinal velocity and the wheel angular velocity in the presence of parameter variations and disturbances using measurements from wheel encoders. An adaptive adjustment of the noise covariances is implemented using a covariance matching technique in the un-scented Kalman filter context for the estimation process. The loss of velocity due to the wheel slip causes extra power consumption. Due to the presence of model uncertainties, parameter variations, and disturbances in the robot nonlinear dynamic system, a sliding mode controller is designed for desired slip control. Experiments are carried out to verify the effectiveness of the estimation algorithm and the controller. In spite of uncertainties presented in the measurements, the robot/wheel dynamics, and terrain condition variations, the controller is able to provide the desired slip ratio control of the mobile robot. It is also demonstrated that the adaptive concept of AUKF leads to better results than the unscented Kalman filter in the robot states estimation which is difficult to measure in practice.