Mini UAV Orientation Control based on Face Tracking Algorithm

This research proposes an orientation control algorithm for unmanned aerial vehicle (UAV) type quadcopter for entertainment uses such as photoshoot and video vlogging. The algorithm consists of face detection, feature extraction, face recognition, and face tracking. There are two experiments designed in this study. The first experiment is used to determine the adaptability of the face recognition algorithm. The second experiment is used to measure the difference between the desired orientation and the actual orientation. The experimental result shows that the proposed algorithm is adaptable. However, there are still several improvements needed such as the recognition performance and orientation accuracy. In conclusion, the current constructed algorithm is promising for further development.