GitHub repository
A robust LiDAR odometry pipeline that works out-of-the-box without parameter tuning for accurate robot localization.
What is GitHub repository?
KISS-ICP is a LiDAR odometry pipeline that estimates a robot's position and orientation in real-time using 3D point cloud data from LiDAR sensors. It solves the problem of accurate robot localization in unknown environments without requiring complex parameter tuning. The system implements a robust point-to-point ICP algorithm that works reliably across different datasets and conditions.
Robotics researchers, autonomous vehicle developers, and engineers building LiDAR-based navigation systems for drones, robots, or mobile platforms. It's particularly valuable for those needing reliable odometry without extensive calibration.
Developers choose KISS-ICP for its simplicity and reliability—it works out-of-the-box without parameter tuning while maintaining accuracy comparable to more complex systems. The clean implementation and ROS 2 integration make it easy to deploy in real robotic applications.
Overview
A LiDAR odometry pipeline that just works
Use Cases
Best For
- Adding LiDAR-based localization to autonomous robots
- Benchmarking and comparing SLAM algorithms
- Educational projects teaching LiDAR odometry fundamentals
- Prototyping robotic navigation systems quickly
- Integrating with ROS 2-based robotic stacks
- Processing outdoor and indoor LiDAR datasets
Not Ideal For
- Applications requiring full SLAM with loop closure and global map optimization
- Teams operating in ROS 1 environments who cannot migrate to ROS 2
- Projects needing multi-sensor fusion (e.g., LiDAR-camera or LiDAR-IMU integration)
Pros & Cons
Pros
Designed to work on most datasets without parameter tuning, as emphasized in the README's tagline 'just works' and the demo showcasing robustness across different conditions.
Adheres to the KISS principle with a minimalistic implementation, making it easy to understand, modify, and integrate into existing systems, as reflected in the clean Python and ROS 2 wrappers.
Provides a dedicated ROS 2 wrapper with clear installation instructions in the README, facilitating straightforward deployment in robotic applications without complex setup.
Based on a peer-reviewed paper from a reputable institution, ensuring the method is validated and reproducible, with citations and reproducible results highlighted in the documentation.
Cons
The README explicitly states ROS 1 is deprecated and not officially supported, forcing users to upgrade to ROS 2, which can be a barrier for legacy robotic systems.
Limited to localization without built-in mapping or loop closure, as indicated by its design as a pipeline, requiring additional modules for complete autonomous navigation.
As a point-to-point ICP method, it inherently accumulates drift over time, a known trade-off that affects long-term accuracy without external correction mechanisms.
Frequently Asked Questions
Related Projects
GitHub repositoryOpen3D: A Modern Library for 3D Data Processing
DracoDraco is a library for compressing and decompressing 3D geometric meshes and point clouds. It is intended to improve the storage and transmission of 3D graphics.
mmdetection3dOpenMMLab's next-generation platform for general 3D object detection.
OpenPCDetOpenPCDet Toolbox for LiDAR-based 3D Object Detection.
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.