EAGLE-VISION-INSPIRED VISUAL MEASUREMENT ALGORITHM FOR UAV’S AUTONOMOUS LANDING

Haibin Duan, Long Xin, Yan Xu, Guozhi Zhao, and Shanjun Chen

Keywords

Unmanned aerial vehicles, autonomous landing, eagle vision, pose estimation

Abstract

In this article, a visual measurement system for an unmanned aerial vehicle (UAV) in an autonomous landing task is implemented. The hardware configuration of the entire platform is introduced on which the visual measurement methods are implemented, and a multicolour-based artificial landing marker is designed. The innovation of this article is that a feature extraction approach inspired by eagle vision mechanism is proposed to detect the landing target. Real-time 6 degree-of-freedom information between the UAV and the landing marker can be obtained from the system. A series of experiments on the UAV platform are conducted to verify the feasibility and effectiveness of the visual measurement system for the landing task, and offline test results compare the proposed eagle feature extraction approach with the traditional method, validating the accuracy, speed and robustness.

Important Links:



Go Back