Proprioceptive State Estimation for Quadruped Robots using Invariant Kalman Filtering and Scale-Variant Robust Cost Functions

Accurate state estimation is crucial for legged robot locomotion, as it provides the necessary information to allow control and navigation. However, it is also challenging, especially in scenarios with uneven and slippery terrain. This paper presents a new Invariant Extended Kalman filter for legged robot state estimation using only proprioceptive sensors. We formulate the methodology by combining recent advances in state estimation theory with the use of robust cost functions in the measurement update. We tested our methodology on quadruped robots through experiments and public datasets, showing that we can obtain a pose drift up to 40% lower in trajectories covering a distance of over 450 m, in comparison with a state-of-the-art Invariant Extended Kalman filter.

I. Introduction

Quadruped robots are becoming increasingly capable, with recent advancements enabling them to perform complex tasks in unstructured environments. State estimation is a crucial step for these robots to achieve autonomy. Quadruped robots require accurate pose and velocity estimation to feed the controllers, ensure stability, and perform navigation. However, accurate state estimation is challenging due to sensor noise and external disturbances.