Cartographer — The Complete Guide: Real-Time SLAM in 2D and 3D
Cartographer is an open-source SLAM library (Apache 2.0) originally developed at Google: it builds a map and locates the robot within it at the same time, in 2D and in 3D. It integrates with ROS and runs on modest onboard hardware.
Cartographer
Cartographer Project
Real-time SLAM in 2D and 3D for robots and mapping platforms.
Get it from the vendorWhat you get
- 2D and 3D SLAM — occupancy grids or 3D maps.
- Sensor fusion — laser (lidar), odometry and IMU.
- Loop closure — correcting drift when the robot returns to a known place.
- Localisation in an existing map — working against a map built earlier.
- Linux and ROS — ready-made packages for integration.
What to know
The original project is no longer actively maintained, and the ROS version is maintained only in a limited capacity. It still works well and is widely used, but do not expect fixes or new features. Tuning matters: map quality depends on the configuration and the sensors. Try it on a recorded dataset or in simulation before the real robot.
Related tools
For the robotics framework itself — ROS. For testing in a simulated environment with sensors — Gazebo or Webots.
ROS (Robot Operating System)
Open Robotics
The standard middleware for building robot software.
Get it from the vendor
Gazebo
Open Robotics
3D robotics simulator with physics and simulated sensors, integrated with ROS.
Get it from the vendor
Webots
Cyberbotics
Approachable robot simulator with a library of ready robots and environments, common in teaching.
Get it from the vendor
CloudCompare
CloudCompare Project
Point cloud and laser scan processing — registration, cleaning, as-built comparison and measurement.
Get it from the vendorThe bottom line
Cartographer is still one of the most mature open-source SLAM tools — just remember it is no longer evolving.