GNIO: Gated Neural Inertial Odometry

TL;DR

GNIO introduces a novel learning-based inertial odometry framework that models motion context and validity, significantly reducing drift and improving accuracy over existing methods through innovative gating and semantic motion querying.

Contribution

The paper presents GNIO, a new framework with a learnable Motion Bank and Gated Prediction Head, enhancing inertial navigation by modeling motion semantics and dynamically suppressing noise.

Findings

60.21% reduction in trajectory error on OxIOD dataset

Outperforms CNN and Transformer baselines in drift reduction

Demonstrates superior generalization in complex motion scenarios

Abstract

Inertial navigation using low-cost MEMS sensors is plagued by rapid drift due to sensor noise and bias instability. While recent data-driven approaches have made significant strides, they often struggle with micro-drifts during stationarity and mode fusion during complex motion transitions due to their reliance on fixed-window regression. In this work, we introduce Gated Neural Inertial Odometry (GNIO), a novel learning-based framework that explicitly models motion validity and context. We propose two key architectural innovations: \ding{182} a learnable Motion Bank that queries a global dictionary of motion patterns to provide semantic context beyond the local receptive field, and \ding{183} a Gated Prediction Head that decomposes displacement into magnitude and direction. This gating mechanism acts as a soft, differentiable Zero-Velocity Update (ZUPT), dynamically suppressing sensor noise during stationary periods while scaling predictions during dynamic motion. Extensive experiments across four public benchmarks demonstrate that GNIO significantly reduces position drift compared to state-of-the-art CNN and Transformer-based baselines. Notably, GNIO achieves a $60.21\%$ reduction in trajectory error on the OxIOD dataset and exhibits superior generalization in challenging scenarios involving frequent stops and irregular motion speeds.