State Estimation for Wheeled-Legged Robot with Slip Detection

The wheeled-legged robot offers enhanced efficiency in exploring tough terrain compared to legged robots. However, difficulties arise when this hybrid locomotion robot operates on slippery terrain, potentially leading to compromised state estimation and control due to the lack of mechanisms to address such conditions. In response to this issue, we propose a framework incorporating an estimated slip ratio into a Kalman Filter (KF) for state estimation to obtain the robot's velocity on slippery terrain. The framework relies solely on data from the Inertial Measurement Units (IMU) and motor sensors mounted on the robot. Through a series of simulations and real-world experiments, the proposed framework is validated on a single-wheeled-legged robot running on slippery terrain, and the results demonstrate its effectiveness in accurately estimating the robot's velocity under slippery conditions.