Visual-servoing and navigation of mobile robots for target tracking applications

摘要

In this paper, we address technical issues involved with development of algorithms and control strategies for vision-based navigation and visual servoing of mobile robots for target tracking applications. The proposed procedural approach includes initial camera calibration, image acquisition, feature extraction, features decision-making. The paper discusses also some of the challenges associated with capturing steady images of moving targets and stabilization of image features. The paper presents our approach to solve the above problems for robust vision-based navigation of mobile robots. The proposed technique has been implemented on a mobile robotic vehicle called "TSU TIGER". TIGER is an all-terrain, four wheeled, differentially driven robotic vehicle with four-wheel independent suspension system and has a variety of navigational sensors. In this paper the implementation issues regarding the proposed methodology on TIGER are discussed.