IONet: Learning to Cure the Curse of Drift in Inertial Odometry

TL;DR

IONet introduces a deep learning approach that segments inertial data into windows and estimates latent states, significantly improving indoor odometry accuracy and generalization to non-periodic motions.

Contribution

The paper presents a novel deep recurrent neural network framework that segments inertial data to accurately estimate odometry, overcoming limitations of traditional integration methods.

Findings

Outperforms state-of-the-art shallow techniques in trajectory estimation

Successfully generalizes to non-periodic motions like trolleys and strollers

Demonstrates robustness across diverse indoor localization scenarios

Abstract

Inertial sensors play a pivotal role in indoor localization, which in turn lays the foundation for pervasive personal applications. However, low-cost inertial sensors, as commonly found in smartphones, are plagued by bias and noise, which leads to unbounded growth in error when accelerations are double integrated to obtain displacement. Small errors in state estimation propagate to make odometry virtually unusable in a matter of seconds. We propose to break the cycle of continuous integration, and instead segment inertial data into independent windows. The challenge becomes estimating the latent states of each window, such as velocity and orientation, as these are not directly observable from sensor data. We demonstrate how to formulate this as an optimization problem, and show how deep recurrent neural networks can yield highly accurate trajectories, outperforming state-of-the-art shallow techniques, on a wide range of tests and attachments. In particular, we demonstrate that IONet can generalize to estimate odometry for non-periodic motion, such as a shopping trolley or baby-stroller, an extremely challenging task for existing techniques.