시각 SLAM 에서의 동적 특징점 검출을 통한 군집 로봇 내의 상대 위치 추정

Abstract

This paper proposes a 3D relative position estimation method for swarm robots through feature-based visual SLAM. Dynamic features are detected and utilized to represent a moving robot. Then, the relative 3D position of the robot is estimated from the depth of the estimated dynamic feature. The proposed algorithm operates in real-time and can be used in various applications.