Vision-based tracking and motion estimation for moving targets using unmanned air vehicles

Abstract

This paper addresses the development of a vision-based target tracking system for a small unmanned air vehicle. The algorithm performs autonomous tracking of a moving target, while simultaneously estimating geographic coordinates, speed, and heading of the target. Tight real-time integration of unmanned air vehicle’s video and telemetry data streams with georeferenced database allows for reliable target identification, increased precision, and shortened time of target motion estimation. A low-cost off-the-shelf system is used, with a modified radiocontrolled aircraft airframe, gas engine, and servos. Tracking is enabled using a low-cost, miniature pan-tilt gimbal. The control algorithm provides rapid target acquisition and tracking capability. A target motion estimator was designed and shown in multiple flight tests to provide reasonable targeting accuracy. The impact of tracking loss events on the control and estimation algorithms is analyzed in detail.