PRBonn/kiss-icp

Health v2 · maintenance only

Adoption not part of the score

2326 stars · 457 forks observed · 2026-09-25

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

KISS-ICP is a LiDAR odometry pipeline built around a simple ICP-based approach that works out of the box on most datasets without parameter tuning. It is written in C++ with Python bindings, a CLI pipeline runner, and ROS 2 support.

Use cases

• estimate vehicle trajectory from lidar point clouds
• run lidar odometry without tuning parameters
• process rosbag lidar data offline
• integrate lidar odometry into a ROS 2 robot stack
• benchmark odometry on SLAM datasets
• build 3D maps from moving lidar scans

When to choose

• you need a simple, dependency-light lidar odometry baseline that works without tuning
• you want a pip-installable CLI to process point cloud data quickly
• you use ROS 2 and need an odometry node
• you need cross-platform (Linux/macOS/Windows) support

When to avoid

• you need full SLAM with loop closure and global mapping
• you rely on ROS 1, which is no longer officially supported
• you need visual-inertial or multi-sensor fusion odometry
• you need highly tuned odometry for extreme conditions like degenerate geometry

Facets

library · maturity active

robotics computer-vision simulation robotics autonomous-vehicles windows python cpp lidar-odometry icp slam point-clouds ros2 3d-mapping algorithms linux macos ros

Member repositories

For agents

markdown · JSON · MCP: product_card(name="PRBonn/kiss-icp")