A ROS-based motion-primitives planner for fast and agile autonomous exploration with aerial robots.
mbplanner_ros is a ROS-based motion-primitives planner for aerial robots, designed to enable fast and agile autonomous exploration in complex environments like caves and mines. It generates dynamically feasible trajectories using pre-computed motion primitives and integrates with mapping frameworks like Voxblox and Octomap for real-time environment representation.
Robotics researchers and developers working on autonomous aerial exploration, particularly those focused on subterranean environments, drone path planning, or ROS-based robotic systems.
It offers a specialized solution for agile exploration with support for multiple mapping backends, simulation environments, and ROS integration, making it suitable for research and development in challenging GPS-denied scenarios.
Motion-primitives Based Planner for Fast & Agile Exploration
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Leverages pre-computed motion primitives to generate dynamically feasible trajectories, enabling fast and agile exploration in complex 3D spaces like caves and mines.
Integrates with both Voxblox (TSDF/ESDF) and Octomap, allowing users to choose the mapping framework based on environment needs, as shown in the configurable launch files.
Includes Gazebo simulations with DARPA Subterranean Challenge cave environments, facilitating testing without physical hardware and supporting realistic scenario development.
Fully integrated with ROS, launchable via launch files and controllable through services, streamlining deployment and interaction in ROS-based robotic systems.
Requires cloning and building the separate mbplanner_ws repository, with additional steps for dependency management, making initial setup time-consuming and error-prone.
Switching between mapping frameworks or modes (e.g., TSDF to ESDF) involves editing header and launch files directly, lacking automated tools and increasing configuration complexity.
The tutorial is hosted externally and may be brief, relying on users to infer details from the README, which lacks comprehensive guides for customization or troubleshooting.