Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

LIO-EKF: High Frequency LiDAR-Inertial Odometry using Extended Kalman Filters

Yibin Wu, Tiziano Guadagnino, Louis Wiesmann, Lasse Klingbeil, Cyrill Stachniss, Heiner Kuhlmann

TL;DR

LIO-EKF targets real-time, high-frequency LiDAR-IMU odometry with minimal tuning by marrying a point-to-point LiDAR front end to a classical error-state EKF back end. The key innovations are an INS-based pose predictor and an adaptive data association threshold that accounts for pose uncertainty, map discretization, and LiDAR noise, enabling reliable correspondences without extensive parameter tuning. Experimental results across urban, campus, and handheld datasets show the method achieves state-of-the-art accuracy while significantly reducing computation time, approaching IMU frame rates. The approach offers a compact, robust alternative to IEKF or factor-graph-based LIO systems, with open-source code to facilitate adoption.

Abstract

Odometry estimation is crucial for every autonomous system requiring navigation in an unknown environment. In modern mobile robots, 3D LiDAR-inertial systems are often used for this task. By fusing LiDAR scans and IMU measurements, these systems can reduce the accumulated drift caused by sequentially registering individual LiDAR scans and provide a robust pose estimate. Although effective, LiDAR-inertial odometry systems require proper parameter tuning to be deployed. In this paper, we propose LIO-EKF, a tightly-coupled LiDAR-inertial odometry system based on point-to-point registration and the classical extended Kalman filter scheme. We propose an adaptive data association that considers the relative pose uncertainty, the map discretization errors, and the LiDAR noise. In this way, we can substantially reduce the parameters to tune for a given type of environment. The experimental evaluation suggests that the proposed system performs on par with the state-of-the-art LiDAR-inertial odometry pipelines but is significantly faster in computing the odometry. The source code of our implementation is publicly available (https://github.com/YibinWu/LIO-EKF).

LIO-EKF: High Frequency LiDAR-Inertial Odometry using Extended Kalman Filters

TL;DR

LIO-EKF targets real-time, high-frequency LiDAR-IMU odometry with minimal tuning by marrying a point-to-point LiDAR front end to a classical error-state EKF back end. The key innovations are an INS-based pose predictor and an adaptive data association threshold that accounts for pose uncertainty, map discretization, and LiDAR noise, enabling reliable correspondences without extensive parameter tuning. Experimental results across urban, campus, and handheld datasets show the method achieves state-of-the-art accuracy while significantly reducing computation time, approaching IMU frame rates. The approach offers a compact, robust alternative to IEKF or factor-graph-based LIO systems, with open-source code to facilitate adoption.

Abstract

Odometry estimation is crucial for every autonomous system requiring navigation in an unknown environment. In modern mobile robots, 3D LiDAR-inertial systems are often used for this task. By fusing LiDAR scans and IMU measurements, these systems can reduce the accumulated drift caused by sequentially registering individual LiDAR scans and provide a robust pose estimate. Although effective, LiDAR-inertial odometry systems require proper parameter tuning to be deployed. In this paper, we propose LIO-EKF, a tightly-coupled LiDAR-inertial odometry system based on point-to-point registration and the classical extended Kalman filter scheme. We propose an adaptive data association that considers the relative pose uncertainty, the map discretization errors, and the LiDAR noise. In this way, we can substantially reduce the parameters to tune for a given type of environment. The experimental evaluation suggests that the proposed system performs on par with the state-of-the-art LiDAR-inertial odometry pipelines but is significantly faster in computing the odometry. The source code of our implementation is publicly available (https://github.com/YibinWu/LIO-EKF).
Paper Structure (18 sections, 16 equations, 2 figures, 4 tables)

This paper contains 18 sections, 16 equations, 2 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Methodology
  4. Error State Model
  5. LiDAR Observation Model
  6. Adaptive Data Association Threshold
  7. Pose Prediction
  8. Map Discretization Errors
  9. Sensor Noise
  10. Experimental Results
  11. Experimental Setup
  12. System Parameters
  13. Odometry Performance
  14. Computation Efficiency
  15. Ablation Studies
  16. ...and 3 more sections

Figures (2)

  • Figure 1: Odometry estimation results of LIO-EKF with different sensing platforms in different environments.
  • Figure 2: Overview of LIO-EKF. We integrate the most basic front end (point-to-point association) and back end (EKF) to build a simple, small yet accurate, generic, and high frequency LIO system.