Drone Swarm Obstacle Avoidance – Collision-Free Navigation of Micro-Drones
Abstract
The Crazyflie micro-drones used at the AerodrOHM do not have dedicated onboard sensors for environmental perception. To enable safe interaction with people in the shared flight space, an external 3D camera is to be installed near the swarm. The goal is a system that reliably detects people in the shared flight space and controls the swarm so that drones automatically avoid and clear the way.
In the first step, the 3D camera is to be integrated into a ROS-based system architecture. The acquired point cloud data must be filtered, pre-processed, and transformed into an appropriate coordinate system. A central aspect is the classification of detected objects: the system must reliably distinguish between people and the drones themselves, using segmentation and 3D object recognition methods such as cluster-based or learning-based approaches.
Building on the person detection, a reactive swarm behaviour is to be developed that ensures safe coexistence in the shared space. This includes defining safety zones around detected persons, dynamically adapting flight paths, and coordinating evasive manoeuvres within the swarm to also avoid collisions between drones. Practical testing takes place in the AerodrOHM flight space.
Work Packages
- Integration and calibration of the external 3D camera
- Processing, filtering and transformation of point cloud data
- Development of a classification to distinguish between persons and drones
- Design and implementation of a reactive swarm behaviour with safety zones
- Test and evaluation in the flight space with real persons
Requirements
- Programming skills (Python and/or C++)
- Ideally some experience with ROS
- Interest in sensor integration, point cloud processing and perception methods
This topic can be completed as a project or master’s thesis subject to agreement.
Supervision
| Role | Name | |
|---|---|---|
| Supervisor | Prof. Dr. Christian Pfitzner | christian.pfitzner@th-nuernberg.de |
Prof. Dr. Christian Pfitzner
Professor of Mobile Robotics
Professor of Mobile Robotics at TH Nürnberg Georg Simon Ohm, Faculty efi. Director of the TTZ Nürnberger Land. Research interests in autonomous navigation, drone technology and sensor fusion.