TY - GEN
T1 - BIPS: Building Information Positioning System
AU - Lee, Max Jwo Lem
AU - Ho, Hiu Yi
AU - Hsu, Li Ta
AU - Au, Stephen Ling Ming
N1 - Funding Information:
This work is supported by PolyU RISUD on the project - BBWK "Resilient Urban PNT Infrastructure to Support Safety of UAV Remote Sensing in Urban Region."
Funding Information:
This work is supported by PolyU RISUD on the project – BBWK “Resilient Urban PNT Infrastructure to Support Safety of UAV Remote Sensing in Urban Region.”
Publisher Copyright:
© 2021 IEEE.
PY - 2021/12
Y1 - 2021/12
N2 - With the rise of digital twins and smart cities, Building Information Modelling have been widely adopted by the construction industry from design, construction to operation maintenance. We present a BIPS (Building Information Positioning System) method which integrates a smartphone VPS (visual positioning system) based on the BIM models, and VO (visual odometry) for the indoor positioning. Firstly, the smartphone images and sensor measurements are sent to a server. In the server, the VPS utilizes computer vision algorithms to extract semantics from the smartphone images. Then, the smartphone image semantics are compared with the BIM semantics. The hypothesized position candidates are distributed in the BIM model. The candidate with the maximum likelihood is regarded as the VPS heading and position estimation. An extended Kalman filter is then used to integrate the VPS with VO, where the former and latter provide measurement and propagation models, respectively. According to the simulation result, the proposed BIPS proves effective in an indoor environment, being capable of improving indoor positioning accuracy to about 1 meter.
AB - With the rise of digital twins and smart cities, Building Information Modelling have been widely adopted by the construction industry from design, construction to operation maintenance. We present a BIPS (Building Information Positioning System) method which integrates a smartphone VPS (visual positioning system) based on the BIM models, and VO (visual odometry) for the indoor positioning. Firstly, the smartphone images and sensor measurements are sent to a server. In the server, the VPS utilizes computer vision algorithms to extract semantics from the smartphone images. Then, the smartphone image semantics are compared with the BIM semantics. The hypothesized position candidates are distributed in the BIM model. The candidate with the maximum likelihood is regarded as the VPS heading and position estimation. An extended Kalman filter is then used to integrate the VPS with VO, where the former and latter provide measurement and propagation models, respectively. According to the simulation result, the proposed BIPS proves effective in an indoor environment, being capable of improving indoor positioning accuracy to about 1 meter.
KW - BIM
KW - Indoor Positioning
KW - Kalman Filter
KW - Semantics
KW - Visual Positioning System
UR - http://www.scopus.com/inward/record.url?scp=85124510394&partnerID=8YFLogxK
U2 - 10.1109/IPIN51156.2021.9662575
DO - 10.1109/IPIN51156.2021.9662575
M3 - Conference article published in proceeding or book
AN - SCOPUS:85124510394
T3 - 2021 International Conference on Indoor Positioning and Indoor Navigation, IPIN 2021
BT - 2021 International Conference on Indoor Positioning and Indoor Navigation, IPIN 2021
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2021 International Conference on Indoor Positioning and Indoor Navigation, IPIN 2021
Y2 - 29 November 2021 through 2 December 2021
ER -
