An open-source 3D LIDAR-based mapping framework for semi-automatic, interactive correction of SLAM mapping failures.
Interactive SLAM is an open-source 3D LIDAR-based mapping framework that enables semi-automatic correction of mapping failures in Simultaneous Localization and Mapping (SLAM). It allows users to interactively fix issues like corrupted odometry, wrong loop detection, and distorted maps, bridging the gap between fully automatic SLAM and manual post-processing. Built on the ROS ecosystem, it integrates with existing SLAM pipelines to produce high-quality, accurate maps.
Robotics researchers, engineers, and developers working on autonomous navigation, 3D reconstruction, or environmental mapping who need precise control over SLAM outputs. It is especially useful for those using LIDAR-based systems and ROS-compatible hardware.
Developers choose Interactive SLAM because it offers a unique blend of automation and user intervention, reducing the time and effort required to correct mapping errors compared to purely automatic SLAM packages. Its intuitive tools and ROS integration make it a practical solution for improving map accuracy in real-world robotics applications.
Interactive Map Correction for 3D Graph SLAM
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Enables manual and automatic fixes for loop closures and map distortions, as shown in features like plane-based map correction and multiple map merging, reducing reliance on purely automatic SLAM.
Built on ROS with compatibility for packages like hdl_graph_slam and LeGO-LOAM, allowing seamless integration into existing robotics workflows and extensibility.
Accepts pose graphs from hdl_graph_slam or odometry from any ROS package, providing versatility in choosing SLAM backends for different use cases.
Offers multiple features including loop closing, plane-based correction, and pose edge refinement, detailed in the key features and examples for thorough map editing.
Requires manual compilation of dependencies like g2o and Ceres, along with specific apt-get commands, making setup time-consuming and prone to errors for newcomers.
Primarily tested on Ubuntu 18.04 with ROS melodic; for kinetic, it only has build-test and requires LLVM toolchain, indicating potential compatibility and stability issues.
The README mentions a new package GLIM has been released with interactive features, suggesting this project might be less maintained or superseded, risking future support.