Attention-Based Neural-Augmented Kalman Filter for Legged Robot State Estimation

TL;DR

This paper introduces an attention-based neural augmentation to the Kalman filter for legged robot state estimation, effectively compensating for slip-induced errors and improving accuracy during slip conditions.

Contribution

It presents a novel neural-augmented Kalman filter with an attention mechanism to estimate and correct slip errors in legged robot state estimation.

Findings

Enhanced state estimation accuracy during slip conditions.

Neural compensator improves robustness over traditional methods.

Attention mechanism effectively models slip severity.

Abstract

In this letter, we propose an Attention-Based Neural-Augmented Kalman Filter (AttenNKF) for state estimation in legged robots. Foot slip is a major source of estimation error: when slip occurs, kinematic measurements violate the no-slip assumption and inject bias during the update step. Our objective is to estimate this slip-induced error and compensate for it. To this end, we augment an Invariant Extended Kalman Filter (InEKF) with a neural compensator that uses an attention mechanism to infer error conditioned on foot-slip severity and then applies this estimate as a post-update compensation to the InEKF state (i.e., after the filter update). The compensator is trained in a latent space, which aims to reduce sensitivity to raw input scales and encourages structured slip-conditioned compensations, while preserving the InEKF recursion. Experiments demonstrate improved performance compared to existing legged-robot state estimators, particularly under slip-prone conditions.