EllipseLIO: Adaptive LiDAR Inertial Odometry with an Ellipsoid Representation

2026-05-20

AI Summary

This work addresses the limited robustness of existing LiDAR-inertial odometry systems across heterogeneous environments and sensor configurations, as well as their reliance on manual parameter tuning. The authors propose a novel adaptive LiDAR-inertial odometry method that, for the first time, employs an ellipsoidal geometric representation to enable adaptive scan filtering and registration without requiring scene-specific parameter adjustments. By integrating tightly coupled LiDAR-IMU fusion with a real-time parameter adaptation mechanism, the approach significantly enhances generalization capability. Experimental evaluation on five diverse datasets demonstrates that the proposed method reduces average odometry error by 38% compared to the next-best approach, with no trajectory divergence observed across all test scenarios.

Abstract

LiDAR Inertial Odometry (LIO) is a critical component for many mobile robots that need to navigate without relying on external positioning (e.g., GPS). Platforms that operate autonomously in different environments and with heterogeneous LiDAR sensors require a LIO approach that can adapt to these different scenarios without human intervention. Existing LIO approaches can typically provide reliable and accurate odometry in scenarios with similar environments and sensors when suitably tuned. However, many approaches struggle to retain robust odometry across heterogeneous environments and sensors while using a consistent configuration. This paper presents EllipseLIO, a real-time LIO approach that generalises between scenarios by using methods for LiDAR scan filtering and registration that adapt to the sensor capabilities and environment without requiring scenario-specific tuning. Experiments with EllipseLIO and state-of-the-art LIO approaches on five datasets with diverse and challenging scenarios demonstrate that EllipseLIO is the best-performing approach overall. It achieves a 38% lower odometry error on average than the second-best approach and is the only approach that does not diverge in any experiment. An open-source version of EllipseLIO will be available at github.com/v4rl-ucy/ellipselio.

Problem

• LiDAR Inertial Odometry
• heterogeneous sensors
• adaptive odometry
• robustness
• environment generalization

Innovation

• adaptive LiDAR inertial odometry
• ellipsoid representation
• sensor-agnostic odometry
• real-time robust localization
• scan registration