Autonomous pursuit unmanned flight system based on location information of a launch-attached location tracker and stereo vision

Abstract

Unmanned aerial vehicles (UAVs) are widely used in both military and civilian applications, and camera-based target tracking is a popular method. However, there is a limitation that cameras alone cannot track targets outside the field of view. To overcome this, we propose an on-board flight system that uses a position tracker with a combined GPS and LTE module and iterative prediction of a Kalman Filter to extend the range of camera-based target tracking. The system consists of three parts: GPS-based pursuit, Stereo Camera-based pursuit, and Obstacle Avoidance, each of which is verified through experiments. Finally, the three parts are integrated into a tracking system, demonstrating that camera-based tracking is possible even when the target is not in the camera's field of view.