cocel-postech/genz-icp

GenZ-ICP: SOTA robust LiDAR odometry (IEEE RA-L 2025)

What it solves

GenZ-ICP addresses the problem of LiDAR odometry instability in degenerate environments—areas where the sensor data lacks sufficient geometric features to accurately determine movement, often leading to drift or failure in localization.

How it works

The system implements a generalizable and degeneracy-robust approach to LiDAR odometry using an adaptive weighting mechanism. It processes LiDAR point clouds to estimate the robot's motion (odometry) while adjusting weights to maintain accuracy even when geometric features are sparse or repetitive.

Who it’s for

This tool is designed for robotics researchers and engineers working on SLAM (Simultaneous Localization and Mapping) and autonomous navigation, particularly those needing reliable localization in challenging environments.

Highlights

  • Degeneracy-Robust: Specifically designed to handle environments where standard ICP (Iterative Closest Point) algorithms struggle.
  • Cross-Platform Support: Available as a Python package via pip and as C++ implementations.
  • ROS Integration: Full support for both ROS 1 and ROS 2 for seamless integration into robotic middleware.
  • Generalizable: Built to be robust across different types of environments without requiring extensive per-scene tuning.

Related

  • Project
    superxslam/SuperOdom

    A lightweight LiDAR-inertial odometry and mapping system that provides robust pose estimation and degeneracy detection for robots using ROS 2.

  • Project
    rsasaki0109/lidar_localization_ros2

    A map-based 3D LiDAR localization system for ROS 2 and Nav2 that uses NDT and GICP algorithms to align sensor scans with 3D point cloud maps.

  • Project
    PRBonn/kiss-icp

    A LiDAR Odometry pipeline that provides accurate and robust registration without requiring manual parameter tuning.

  • Project
    liangheming/FASTLIO2_ROS2

    A ROS2 adaptation of FASTLIO2 that adds loop closure, relocalization, and consistency map optimization for LiDAR-Inertial Odometry.

  • Project
    ouguangjun/Leg-KILO

    A real-time LiDAR-based SLAM system designed for high-dynamic platforms like legged robots, featuring a two-stage ESKF frontend and factor graph backend for robust mapping.