Monocular Vision-Based Collision-Free Navigation for Flying Robots
Journal
Proceedings - 2019 8th International Congress on Advanced Applied Informatics, IIAI-AAI 2019
Pages
1091-1092
Date Issued
2019
Author(s)
Fu, J.-S.
Abstract
In this paper, a systematic navigation structure for flying robots is proposed. In particular, the large scale direct SLAM (LSD-SLAM) is implemented to provide self-localization and semi-dense depth maps. The collected depth maps are then integrated into a three-dimensional occupancy map with the Octomap, and the occupancy information of the currently built map is utilized for the navigation of the flying robot. Dealing with the scale recovery problem of monocular vision-based systems, such as the LSD-SLAM system, an estimate of the scale is calculated with the aid of an ultrasound sensor. Given the estimated scale, the flying robot can then be navigated with the re-scaled pose provided by the LSD-SLAM system. The proposed system avoids collision of the robot with any obstacle, and the paths are only generated within the area free of collision. Experiments in both simulation and the real world are provided to verify the proposed system.
Type
conference paper