hku-mars/FAST-LIVO2

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

4557 stars · 844 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

FAST-LIVO2 is a fast, tightly-coupled LiDAR-inertial-visual odometry and mapping system written in C++ on ROS. It provides real-time, accurate localization and colored 3D reconstruction, suitable for onboard robotic use in degraded environments.

Use cases

• real-time lidar-inertial-visual odometry for drones and robots
• colored 3D point cloud mapping from lidar and camera
• localization in degraded environments like tunnels or darkness
• building datasets for NeRF or gaussian splatting reconstruction
• onboard SLAM for resource-constrained platforms
• mesh reconstruction from fused sensor data

When to choose

• you need accurate real-time odometry fusing lidar, IMU, and camera
• you want colored point clouds or 3D reconstruction output from a SLAM system
• you run ROS on Ubuntu with a supported Livox or similar lidar setup
• you need onboard localization on resource-constrained hardware

When to avoid

• you need a pure visual SLAM without lidar
• you are not using ROS or Linux
• you need a plug-and-play solution without sensor calibration and synchronization
• your project requires a permissive license (GPL-2.0)

Facets

library · maturity active

robotics computer-vision graphics simulation sdk robotics autonomous-vehicles computer-vision simulation cpp slam lidar odometry sensor-fusion 3d-reconstruction point-cloud nerf gaussian-splatting localization mapping algorithms linux ros

Member repositories

For agents

markdown · JSON · MCP: product_card(name="hku-mars/FAST-LIVO2")