hku-mars/Point-LIO

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1318 stars · 213 forks observed · 2026-08-28

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

Point-LIO is a robust high-bandwidth LiDAR-inertial odometry framework that estimates ego-motion and builds maps by fusing LiDAR point clouds with IMU data. It outputs odometry at very high frequencies (4k-8kHz) and remains robust under aggressive motions, IMU saturation, and severe vibration.

Use cases

• estimate robot trajectory from LiDAR and IMU data
• build 3D maps while a drone or robot moves aggressively
• get high-frequency odometry for real-time feedback control
• run odometry on datasets with fast rotations and severe vibration
• perform LiDAR-inertial odometry without an IMU using a gravity prior
• support trajectory planning and perception with distortion-free point clouds

When to choose

• your platform experiences high angular/linear velocities or vibration that breaks standard LIO methods
• you need odometry output at kHz rates for control or perception loops
• your IMU saturates during aggressive motion and you need robustness to that
• you want per-point processing to avoid motion distortion in mapping

When to avoid

• you need a full SLAM system with loop closure rather than pure odometry
• your LiDAR point clouds lack per-point timestamps, which Point-LIO requires
• you need a plug-and-play solution without tuning IMU saturation and extrinsic parameters
• you work outside ROS since the tooling is ROS-centric

Facets

library · maturity active

robotics simulation computer-vision robotics autonomous-vehicles computer-vision cpp lidar odometry slam imu point-cloud state-estimation kalman-filter ros sensor-fusion algorithms linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="hku-mars/Point-LIO")