Key points are not available for this paper at this time.
Automated Localization of UAVs in GPS-Denied Indoor Construction Environments Using Fiducial Markers Mohammad Nahangi, Adam Heins, Brenda McCabe and Angela Schoellig Pages 88-94 (2018 Proceedings of the 35th ISARC, Berlin, Germany, ISBN 978-3-00-060855-1) Abstract: Unmanned Aerial Vehicles (UAVs) have opened a wide range of opportunities and applications in different sectors including construction. Such applications include: 3D mapping from 2D images and video footage, automated site inspection, and performance monitoring. All of the above-mentioned applications perform well outdoors where GPS is quite reliable for localization and navigation of UAVs. Indoor localization and consequently indoor navigation have remained relatively untapped, because GPS is not sufficiently reliable and accurate in indoor environments. This paper presents a method for localization of aerial vehicles in GPS - denied indoor construction environments. The proposed method employs AprilTags that are linked to previously known coordinates in the 3D building information model (BIM). Using cameras on-board the UAV and extracting the transformation from the tag to the cameras frame, the UAV can be localized on the site. It can then use the previously computed information for navigation between critical locations on construction sites. We use an experimental setup to verify and validate the proposed method by comparing with an indoor localization system as the ground truth. Results show that the proposed method is sufficiently accurate to perform indoor navigation. Moreover, the method does not intensify the complexity of the construction execution as the tags are simply printed and placed on available surfaces at the construction site. Keywords: Indoor localization, construction performance monitoring, Building Information Model (BIM), AprilTags, Automated monitoring, Facility Management, DOI: https://doi.org/10.22260/ISARC2018/0012 Download fulltext Download BibTex Download Endnote (RIS) TeX Import to Mendeley
Nahangi et al. (Sun,) studied this question.