Please use this identifier to cite or link to this item:
Title: Autonomous elevator inspection with unmanned aerial vehicle
Authors: Kit, HT 
Chen, H 
Keywords: Computer Vision
Indoor Navigation
Issue Date: 2017
Publisher: Institute of Electrical and Electronics Engineers
Source: 2016 Asia-Pacific World Congress on Computer Science and Engineering and Asia-Pacific World Congress on Engineering, APWC on CSE/APWCE 2016, Sofitel Fiji Resort and SpaNadi, Fiji, 4 - 6 December 2016, 7941936, p. 26-33 How to cite?
Abstract: Hong Kong has the highest elevator density in the world, with more than 63,000 elevators. The inspection and maintenance of elevator rope have been carried out directly by human with simple inspection gadgets traditionally. Recently however, the big gap between dramatically increasing demand of elevator inspection and insufficient manpower in the related industry, as well as the safety issues coupled with human inspection have resulted in investigations of the possibility in adopting Unmanned Aerial Vehicle (UAV) for elevator rope inspection. Nevertheless, UAV's control and navigation in a relatively small (comparing to the size of UAV itself) elevator shaft with extremely low luminance and high air turbulence are the biggest challenges in this solution, especially when its bottom camera, which serves as a crucial input for hovering, is nearly unplugged. This project intends to propose and develop an effective methodology of elevator rope inspection with an UAV. Vision-based control and navigation algorithms for facilitating the UAV with autonomous elevator rope detection, tracking, and inspection have been proposed. Simulations and experiments were conducted with a real setup of elevator shaft. Experimental results verified the effectiveness and efficiency of the proposed method. The proposed method can address the above issues confronted by the elevator maintenance service providers.
ISBN: 9781509057535
DOI: 10.1109/APWC-on-CSE.2016.016
Appears in Collections:Conference Paper

View full-text via PolyU eLinks SFX Query
Show full item record

Page view(s)

Citations as of Nov 12, 2018

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.