Wheel Slippage Detection Using IMU and Classification for a Wheeled Mobile Robot

Abstract

In this paper a novel approach for wheel slippage detection in wheeled mobile robots is explored. The core idea is to classify data from the inertial measurement unit (IMU) to detect wheel slippage for a differential drive mobile platform. The developed mobile platform was fitted with an inertial measurement unit, was used to gather data under various scenarios. The collected data was subsequently analyzed and labeled for the training multiple. The performance of these classifiers was investigated, and the results are presented. The most effective model was then deployed on a low-level microcontroller for further real-time testing.