Reconstructing a 3D room from a Kinect carrying UAV

This project focuses on the creation of a portable SLAM (Simultaneous Localisation and Mapping) system, which uses an Unmanned Aerial Vehicle (UAV) as the transportation medium. The main purpose of the system is to create a 3D map of the environment, while concurrently localizing itself within the map. The real world applications of this system concentrate on search and rescue scenarios. The system uses the Microsoft Kinect as its primary sensor. Within this project we utilized Visual SLAM, which is the process of using data from the Kinect sensor to calculate position. The algorithm looked at successive frames and depth estimates from the Kinect and then matched features across the images to calculate distance and stability. The work presented in this paper is approached from a practical point of view rather than purely theoretical basis. The end result is a physical prototype which is ready to be deployed in the field for further testing.