Tutorials

Tutorials of RTAB-Map. Installation instructions can be found here.

Robot RGB-D mapping (ROS)

If you want to use RTAB-Map with ROS, the easiest way is to use the ros-pkg provided, refer to [ROS#Demos ROS Demos]. Otherwise, you could use the RTAB-Map library in C++, giving the required inputs (see this [RGBDExample example] of usage).

Robot configurations

• This page shows how to integrate RTAB-Map on your robot.

Standalone RTAB-Map application

Hand-held Kinect RGB-D Mapping

• This tutorial shows how to use RTAB-Map out-of-the-box to map an environment using a hand-held Kinect.

Fast RGB-D Mapping

• This tutorial shows how to setup RTAB-Map with fast odometry (~20/30Hz).

Multi-session mapping

• This tutorial shows a demonstration with real robot data for multi-session mapping.

Appearance-based loop closure detection

• This simple example shows how to process your own data set (images-only) with RTAB-Map.

Benchmark loop closure detection

• This tutorial shows how to use rtabmap-console tool to process all data sets in the paper "Appearance-Based Loop Closure Detection for Online Large-Scale and Long-Term Operation", then using MATLAB and ground truth files to show Precision/Recall curves.

Plot statistics

• This tutorial will show the power of GUI for online monitoring of the core process.

C++ usage (library)

How to use RTAB-Map library in your C++ project.

• C++ usage example for RGB-D Mapping
• C++ usage example for appearance-based loop closure detection only