위성체 근접 운용기법을 위한 스테레오 비전 알고리즘

Abstract

It is very important to estimate the relative positions of the satellites. Estimation methods of relative distances between satellites are proposed based on stereo visions in this work. Correspondence problems are solved by analyzing the feature points existing in two images. A homography matrix of the system is generated based on the matched feature points. It also distinguished the unconformed feature points by the RANSAC (Random Sample Consensus) algorithm. It is confirmed that the accuracy of the proposed algorithm for position estimations between the targets can be guaranteed with comparatively small numbers of feature points.