Data-Driven Slip Model for Improved Localization and Path Following applied to Lunar Micro-Rovers

Abstract

Micro-lunar rovers need to solve a slew of challenges on the Moon, with no human intervention. One such challenge is the need to know their location in order to navigate and build maps. However, localization is challenging on the moon due to lack of supporting infrastructure, such as global positioning satellites. It is made more difficult due to constraints on available sensing and computing on such rovers. Low gravity and regolith properties cause varying wheel slippage, further complicating position estimation.

This thesis presents a data-driven slip model comprising of 3 layers that enables lunar micro rovers to perform more accurate position estimation through wheel odometry and follow planner commanded drive-arc closely. This is crucial in micro rovers with sensors limitations, arising from the inability to carry better sensors due to size and weight limitations. We start by describing a kinematics model formulation leveraged from previous literature which allows easy integration of our slip model. Next, we describe in detail the layers in the slip model and present how the parameters of the model are derived. We conclude by demonstrating the effectiveness of our slip model in achieving our localization and controls goals through testing using a skid-steer surrogate rover.