Autonomous Landing and Charging, a prototype project

The CopterSonde team was supported by a CIWRO Director’s Discretionary Research Fund grant to develop a prototype system that could support the rapid conceptual trialing of automated landing and charging of UAS. In the early 2023, the CopterSonde team focused on market research before selecting a potential charging system. As the 2023-2024 academic year got started, the research group teamed up with a Capstone research team from OU’s Aerospace and Mechanical Engineering department. The fours students spent the year working alongside the research team on research and development tasks.

At then end of the year, the collaboration resulting in an extension of a commercial-of-the-shelf charging pad, quadrupling the effective area for charging. An IR-based beacon was integrated into the charging pad to support autolanding, and the entire system was mounted onto a trailer-based platform for transport. The student team built a test-UAS for trialing the development systems (wouldn’t want to crash an operational CopterSonde, would we!?!), which included integrating charging and auto-landing functionality. Early tests were successful. Autonomous charging capabilities have been demonstrated without the need for precise, repeatable landing location on the charging surface. Any two contact points will work. IR-based auto-landing was also demonstrated, but with mixed success. The metalic surface of the charging pad and even the reflections from surrounding surfaces produced enough interference with the beacon’s signal to confuse the landing detection system. Specialized paints can mitigate some of these issues. Other mitigation steps will be explored in future development. This project was also able to demonstrate advancement in software required for a system like this to work: the charging system must recognize the drone to provide the right power to it; the launch sequence must include the current charge state and weather conditions in its decision sequence; the landing sequence must recognize if charging begins to determine if landing is successful. These and other integrations require cross-system communication. At the end of the project, the team was able to demonstrate a remotely launched profile flight sequence via this software successfully.

This project has a lot of room to grow in the future, and is intended to serve as an early prototype where researchers can try techniques and tools easily and at low cost. Overall, this project was a large success at relatively low cost.