Light Mode

AirIO: Learning Inertial Odometry with
Enhanced IMU Feature Observability

Yuheng Qiu*1, Can Xu*1, Yutian Chen1, Shibo Zhao1, Junyi Geng2 and Sebastian Scherer1
Carnegie Mellon University
GitHub RepoarXiv PageSupplemental Materials

* Equal Contribution.

1 Robotics Institute, Carnegie Mellon University.

2 Department of Aerospace Engineering, Pennsylvania State University.

Abstract

Inertial odometry (IO) using only Inertial Measurement Units (IMUs) offers a lightweight and cost-effective solution for Unmanned Aerial Vehicle (UAV) applications, yet existing learning-based IO models often fail to generalize to UAVs due to the highly dynamic and non-linear-flight patterns that differ from pedestrian motion. In this work, we identify that the conventional practice of transforming raw IMU data to global coordinates undermines the observability of critical kinematic information in UAVs. By preserving the body-frame representation, our method achieves substantial performance improvements, with a 66.7% average increase in accuracy across three datasets. Furthermore, explicitly encoding attitude information into the motion network results in an additional 23.8% improvement over prior results. Combined with a data-driven IMU correction model (AirIMU) and an uncertainty-aware Extended Kalman Filter (EKF), our approach ensures robust state estimation under aggressive UAV maneuvers without relying on external sensors or control inputs. Notably, our method also demonstrates strong generalizability to unseen data not included in the training set, underscoring its potential for real-world UAV applications.

Introduction

Results

Without external sensors or control information, AirIO achieves up to a 86.6% performance boost over SOTA methods

Interactive Demo

Demo may take several seconds to load, depending on your network connection status.

AirIO on EuRoC MH04/rerun/euroc_mh04.rrd

Methods

We identify the commonly used global-coordinate approach is suboptimal for UAVs due to their dynamic nature. Two simple steps to achieve significant gains:

  1. Predicting velocity using body-coordiante frame representation.
  2. Explicitly encoding UAV attitude information.

Figure 1. Blackbird
Figure 2. EuRoC
Figure 3. Pegasus

System Pipeline

Color Inversion ON
Figure 4. By integrating the novel AirIO network and an uncertainty-aware IMU preintegration model into an EKF, we achieve robust odometry even under aggressive maneuvers.