Low-Cost Monocular Vision-Based Localization and Feedback Control of UAS for Classroom and Education Settings

We introduce a cost-effective real-time 3-dimensional localization and control system of a quad rotorcraft Unmanned Aircraft System (UAS). The intended purpose of the system is to be implemented in low-budget classrooms to teach real-time closed-loop control of UASs and expand interest in the field. The system's localization technique only requires light-emitting diode (LED) markers and a conventional webcam device. Relying on the sampled visual location data, a closed-loop feedback control is developed for real-time stabilization. The system can be implemented with less than 1% of the monetary cost of a conventional motion capture system (MCS), which allows more institutions to purchase and use the materials. To validate our approach, the positional data from the proposed localization system is compared to ground truth data from a conventional MCS. Results from several real-time experiments are provided to validate the reliability and applicability of the localization system in UAS control.