A robust and efficient trajectory planner enabling quadrotor fast flight in complex unknown environments.
Fast-Planner is an open-source trajectory planning framework for quadrotors that enables fast, autonomous flight in complex and unknown environments. It solves the problem of generating safe, dynamically feasible, and efficient trajectories by combining kinodynamic path searching, B-spline optimization, and topological exploration. The framework provides a modular codebase that supports real-time replanning and integrates with volumetric mapping for obstacle avoidance.
Researchers and developers in aerial robotics, particularly those working on autonomous drone navigation, motion planning, and quadrotor control systems. It is also suitable for academic projects and open-source drone initiatives requiring robust planning algorithms.
Developers choose Fast-Planner for its proven robustness, computational efficiency, and comprehensive suite of planning algorithms tailored for high-speed flight. Its modular design and support for popular drone projects make it a versatile foundation for both research and practical applications.
A Robust and Efficient Trajectory Planner for Quadrotors
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Implements robust kinodynamic path searching and B-spline optimization, validated through IEEE RA-L and ICRA papers for safe, efficient trajectories.
Generates multiple topologically distinct paths to avoid local minima, enhancing navigation robustness in complex 3D environments as demonstrated in simulations.
Serves as a base for popular open-source projects like ego-planner and FUEL, facilitating extensibility and reuse in research and development.
Designed for computational efficiency with online ESDF mapping and fast trajectory generation, enabling autonomous flight in unknown settings.
Perception-aware planning is listed as 'to appear' in the README, limiting functionality for active obstacle avoidance until future updates.
Requires specific ROS versions, nlopt v2.7.1 compilation, and optional CUDA for GPU rendering, making initial configuration error-prone.
Released under GPLv3 with a disclaimer offering no warranties, lacking commercial support or certifications for production deployments.