Adaptive Autonomous Landing System for Quadcopter Drones Through Descent Assisted by Visual Recognition

Cutipa J.; Lizarbe D.; Nieves A.; Carrizales J.; Sampen L.; Becerra J. · Proceedings of the 2025 IEEE 32nd International Conference on Electronics, Electrical Engineering and Computing, INTERCON 2025 (2025) · DOI: 10.1109/intercon67304.2025.11244630

Resumen

Landing is an essential phase in the operation of drones or unmanned aerial vehicles (UAVs), especially when maneuvers are performed by untrained or inexperienced pilots. Irregular and imprecise maneuvers and inclined or irregular surfaces, during landing, result in abrupt landings, rollovers, or damage to the drone. This problem is more prevalent in repetitive missions and uncontrolled environments where the precision of landing the drone is fundamental. This project proposes the design and development of an adaptive autonomous landing system for quadcopter drones. The descent to the platform will be made by implementing an assisted landing technique based on ArUco marker detection. This allows us to calculate the relative position and orientation between the drone and the platform with the ArUco marker and thus designate the route for a safe descent. In addition, the drone will have a landing gear that allows it to land on slightly inclined surfaces. The landing gear will regulate the position of its legs according to the inclination of the drone with respect to the platform with the ArUco marker. The final result of the article is that the drone will make a precise and safe landing on the platform with the ArUco marker, which will be on an inclined surface, thus reducing the damage caused by abrupt landings. The validation of the results will be done through 3D simulations using Gazebo ROS and physical tests.

Instituciones: Universidad Peruana de Ciencias Aplicadas