vectr-ucla/direct_lidar_inertial_odometry

Health v2 · maintenance only

Adoption not part of the score

1039 stars · 260 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

DLIO is a lightweight LiDAR-inertial odometry algorithm that constructs continuous-time trajectories using a coarse-to-fine approach for precise motion correction. It is a C++/ROS package supporting common LiDARs (Ouster, Velodyne, Hesai, Livox) and was published at IEEE ICRA 2023.

Use cases

• estimate robot trajectory from LiDAR and IMU data
• build 3D maps while localizing a moving robot
• perform real-time odometry on resource-constrained robots
• correct motion distortion in LiDAR point clouds
• run SLAM-style localization with a Livox or Ouster sensor
• replace heavier LIO pipelines like LIO-SAM with a lighter alternative

When to choose

• you need lightweight, real-time LiDAR-inertial odometry in a ROS stack
• your robot has a 6-axis IMU and a supported LiDAR (Ouster, Velodyne, Hesai, Livox)
• you need precise motion correction for fast-moving platforms
• you want an MIT-licensed, research-backed odometry algorithm

When to avoid

• you need visual or visual-inertial SLAM with cameras
• your LiDAR and IMU cannot be time-synchronized
• you need a pure LiDAR solution without an IMU
• you are not using ROS or Linux

Facets

library · maturity active

simulation computer-vision middleware robotics autonomous-vehicles cpp lidar-inertial-odometry slam lidar imu odometry mapping localization continuous-time-trajectory ros point-clouds algorithms linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="vectr-ucla/direct_lidar_inertial_odometry")