SE(3)-LIO: Smooth IMU Propagation with Jointly Distributed Poses on SE(3) Manifold for Accurate and Robust LiDAR-Inertial Odometry
Gunhee Shin, Seungjae Lee, Jei Kong, Young-Woo Seo, Hyun Myung
AI summary
Problem
Existing IMU propagation methods separate rotation and translation, failing to account for rotational effects on translation and causing inaccurate motion prediction. Furthermore, motion compensation typically ignores the uncertainty and correlation between predicted poses, leading to distorted measurements and degraded odometry.
Approach
The method propagates 6-DoF poses jointly on the SE(3) manifold to capture rotational variation, and derives a joint distribution of predicted poses to accurately quantify relative transformation uncertainty for probabilistic motion compensation.
Key results
- SE(3) manifold propagation incorporates rotational variation into translation for accurate motion prediction
- Uncertainty-aware motion compensation models relative transformation uncertainty using jointly distributed poses
- SE(3)-LIO integrates these components into an error-state Kalman filter framework
- Demonstrated superior accuracy and robustness over state-of-the-art methods on aggressive and large-scale datasets
Why it matters
Enables reliable autonomous navigation in dynamic, high-speed, or large-scale environments where traditional odometry methods struggle.
Abstract
In estimating odometry accurately, an inertial measurement unit (IMU) is widely used owing to its high- rate measurements, which can be utilized to obtain motion information through IMU propagation. In this paper, we address the limitations of existing IMU propagation methods in terms of motion prediction and motion compensation. In motion prediction, the existing methods typically represent a 6-DoF pose by separating rotation and translation and propagate them on their respective manifold, so that the rotational variation is not effectively incorporated into translation propagation. During motion compensation, the relative transformation be- tween predicted poses is used to compensate motion-induced distortion in other measurements, while inherent errors in the predicted poses introduce uncertainty in the relative transfor- mation. To tackle these challenges, we represent and propagate the pose on SE(3) manifold, where propagated translation properly accounts for rotational variation. Furthermore, we precisely characterize the relative transformation uncertainty by considering the correlation between predicted poses, and incorporate this uncertainty into the measurement noise during motion compensation. To this end, we propose a LiDAR-inertial odometry (LIO), referred to as SE(3)-LIO, that integrates the proposed IMU propagation and uncertainty-aware motion compensation (UAMC). We validate the effectiveness of SE(3)- LIO on diverse datasets. Our source code and additional material are available at: https://se3-lio.github.io/.