explaingit

introlab/rtabmap

★ 3,768C++Audience · researcherComplexity · 4/5LicenseSetup · hard

TLDR

RTAB-Map is a C++ robotics library that lets a robot or camera-equipped device figure out where it is and build a map of its surroundings simultaneously, a technique called SLAM. Works with ROS or as a standalone app on desktop, Android, and iOS.

Mindmap

mindmap
  root((rtabmap))
    What it does
      SLAM mapping
      Localization
      Place recognition
    Tech Stack
      C++
      ROS 1 and 2
      Docker
    Use Cases
      Autonomous robots
      Drone navigation
      3D environment maps
    Audience
      Robotics researchers
      Systems engineers
Click or tap to explore — scroll the page freely

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

Things people build with this

USE CASE 1

Build a robot that maps an unknown room while navigating it at the same time.

USE CASE 2

Add real-time position tracking to an autonomous drone or vehicle.

USE CASE 3

Create a 3D map of a physical space using a handheld depth camera.

USE CASE 4

Integrate visual place-recognition into a ROS-based robotics project.

Tech stack

C++ROSROS 2DockerAndroidiOS

Getting it running

Difficulty · hard Time to first run · 1h+

Requires a working ROS installation and a compatible depth or stereo camera, GPU acceleration is optional but recommended for real-time performance.

Use, modify, and distribute freely including for commercial purposes, as long as you retain the copyright notice. BSD license.

In plain English

RTAB-Map is a C++ library and standalone application for mapping and localization used in robotics. The name stands for Real-Time Appearance-Based Mapping. In practical terms, it lets a robot or a device with a camera figure out where it is in a space and build a map of that space at the same time, a task the robotics field calls SLAM (simultaneous localization and mapping). As the robot moves, RTAB-Map recognizes previously seen places by their visual appearance and uses that to correct its position estimate over time. The library works as a component inside ROS, the Robot Operating System, which is a widely used framework in academic and commercial robotics. It also runs without ROS as a standalone app, and it supports Android and iOS through the listed topics. It can be installed through standard ROS package repositories for both ROS 1 and ROS 2, or pulled as a Docker image. The project comes from IntRoLab, a robotics research lab at the University of Sherbrooke in Quebec, Canada. The README is brief and points to a project website and a wiki for installation instructions, usage examples, and links to academic papers. The code is released under a BSD license. If you are not working in robotics or autonomous systems, this library is unlikely to be relevant. It is a research tool built for people building robots or autonomous devices that need to navigate physical spaces.

Copy-paste prompts

Prompt 1
I want to use RTAB-Map with ROS 2 to build a map of my apartment using a RealSense depth camera. Walk me through the setup steps and the launch commands I need.
Prompt 2
Using RTAB-Map's standalone app on Android, how do I record a 3D map of a room and export it as a point cloud?
Prompt 3
I'm building an autonomous robot in ROS. How do I configure RTAB-Map so it corrects its location estimate when it revisits a place it has already seen?
Prompt 4
Show me how to pull the RTAB-Map Docker image, run it with a live camera feed, and visualize the resulting 3D map.
Open on GitHub → Explain another repo

← introlab on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.