Neural Aided Adaptive Innovation-Based Invariant Kalman Filter

arXiv cs.RO / 3/31/2026

💬 OpinionIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper introduces a neural-aided adaptive invariant Kalman filter that performs innovation-based process noise adaptation within a Lie-group (tangent Lie space) framework rather than a standard Euclidean one.
It derives a theoretical extension for process noise estimation formulated directly on Lie groups and pairs it with a lightweight neural network that predicts process-noise covariance parameters from raw inertial data.
The neural component is trained in a sim2real setup using domain adaptation, aiming to learn motion- and sensor-dependent noise characteristics without labeled real-world data.
Experiments on autonomous underwater navigation show improved position root mean square error versus existing approaches, suggesting geometric invariance helps learning-based adaptation.
Overall, the work claims that combining geometric invariance with tangent-space adaptive noise estimation yields a robust method for more accurate localization in nonlinear autonomous platforms.

Abstract

Autonomous platforms require accurate positioning to complete their tasks. To this end, a Kalman filter-based algorithms, such as the extended Kalman filter or invariant Kalman filter, utilizing inertial and external sensor fusion are applied. To cope with real-world scenarios, adaptive noise estimation methods have been developed primarily for classical Euclidean formulations. However, these methods remain largely unexplored in the tangent Lie space, despite it provides a principled geometric framework with favorable error dynamics on Lie groups. To fill this gap, we combine invariant filtering theory with neural-aided adaptive noise estimation in real-world settings. To this end, we derive a novel theoretical extension of classical innovation-based process noise adaptation formulated directly within the Lie-group framework. We further propose a lightweight neural network that estimates the process noise covariance parameters directly from raw inertial data. Trained entirely in a sim2real framework via domain adaptation, the network captures motion-dependent and sensor-dependent noise characteristics without requiring labeled real-world data. To examine our proposed neural-aided adaptive invariant Kalman filter, we focus on the challenging real-world scenario of autonomous underwater navigation. Experimental results demonstrate superior performance compared to existing methods in terms of position root mean square error. These results validate our sim2real pipeline and further confirm that geometric invariance significantly enhances learning-based adaptation and that adaptive noise estimation in the tangent Lie space offers a powerful mechanism for improving navigation accuracy in nonlinear autonomous platforms.