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

2026-01-26

IEEE Robotics and Automation Letters

AI Summary

This work addresses the state estimation bias in legged robots caused by foot slippage, which violates the no-slip assumption commonly adopted in traditional estimators. To mitigate this issue, the authors propose a neural-augmented Invariant Extended Kalman Filter (InEKF) that incorporates an attention mechanism following the InEKF update step. A neural compensator is trained in a latent space to dynamically correct state estimation errors based on the severity of slippage. By integrating the recursive structure of the InEKF, attention-driven contextual awareness, and the nonlinear compensation capability of neural networks, the proposed framework significantly outperforms existing methods in slippery environments, achieving higher accuracy and robustness in state estimation.

Technology Category

• Intelligent Robots: State Estimation
• Machine Learning: Calibration & Uncertainty Quantification
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Agentic search

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.

Problem

• legged robot
• state estimation
• foot slip
• Kalman filter
• estimation error

Innovation

• Attention mechanism
• Neural-Augmented Kalman Filter
• Legged robot state estimation
• Foot slip compensation
• Invariant Extended Kalman Filter