Visual-Inertial Indoor Navigation Systems and Algorithms for UAV Inspection Vehicles

Abstract

In UAV navigation, one of the challenges in which considerable efforts are being focused is to be able to move indoors. Completing this challenge would imply being able to respond to a series of industrial market needs such as the inspection of internal environments for safety purpose or the inventory of stored material. Usually GPS is used for navigation, but in a closed or underground environment, its signal is almost never available. As a consequence, to achieve the goal and ensure that the UAV is able to accurately estimate its position and orientation without the usage of GPS, an alternative navigation system based on visual-inertial algorithms and the SLAM will be proposed using data fusion techniques. In addition to the navigation system, we propose an obstacle avoidance method based on a Lidar sensor that allows navigation even in the absence of light.