RTAB-Map is a simultaneous localization and mapping library and standalone application for robotics.
The tool addresses the problem of enabling robots to build maps of their environment while simultaneously determining their position within those maps. RTAB-Map uses appearance-based loop closure detection to correct drift in trajectory estimation, allowing it to produce consistent maps even over long exploration sequences. The system operates as both a C++ library that can be integrated into larger applications and as a standalone tool, with native support for ROS and ROS2 integration.
The tool suits robotics projects that require robust mapping and localization, particularly those where loop closure detection is valuable for maintaining map consistency. It runs on Linux, Windows, and macOS, and has been ported to mobile platforms including Android and iOS. The project provides a C++ API with comprehensive parameter documentation and command-line tools for offline processing and analysis. Those adopting the tool should expect to work with ROS integration or use the library directly in C++ applications, and should consult the wiki for installation guidance and usage examples.
The project maintains a substantial user base, with almost all open issues originating from outside users rather than the core team. Maintainers typically respond to new issues and pull requests within a few days. Work in the issue tracker is dominated by enhancement requests, iOS-related issues, and bug reports.
