Transfer learning enhanced AR spatial registration for facility maintenance management
Augmented reality (AR), which requires a spatial registration technique, has proved to greatly improve the efficiency of facility maintenance management (FMM) activities. Being one of the most promising techniques for indoor localization, Wi-Fi fingerprinting has been widely used for AR spatial regi...
| Main Authors: | , , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/161072 |
| _version_ | 1811682950569263104 |
|---|---|
| author | Chen, Keyu Yang, Jianfei Cheng, Jack C. P. Chen, Weiwei Li, Chun Ting |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Chen, Keyu Yang, Jianfei Cheng, Jack C. P. Chen, Weiwei Li, Chun Ting |
| author_sort | Chen, Keyu |
| collection | NTU |
| description | Augmented reality (AR), which requires a spatial registration technique, has proved to greatly improve the efficiency of facility maintenance management (FMM) activities. Being one of the most promising techniques for indoor localization, Wi-Fi fingerprinting has been widely used for AR spatial registration. However, localization accuracy of Wi-Fi fingerprinting decreases over time due to dynamics of environmental factors. Readings from different mobile devices can also affect the accuracy negatively. In this paper, a transfer learning technique named transferable CNN-LSTM is proposed for improving the robustness of Wi-Fi fingerprinting while implementing AR in FMM activities. Convolutional neural network (CNN), embedded with long short term memory (LSTM) networks, is utilized to predict the location of unlabeled fingerprints. Multiple kernel variant of maximum mean discrepancy (MK-MMD) is adopted to reduce the distribution difference between the source domain and the target domain, so that the location of the newly collected unlabeled fingerprints can be predicted accurately. As shown in the experimental validation, the transferable CNN-LSTM can achieve an accuracy of 97.1% in short-term (without significant environmental changes) spatial registration, 87.8% in long-term (with significant environmental changes) spatial registration, and around 90% in multi-device spatial registration, indicating a higher accuracy and better robustness over other conventional approaches. |
| first_indexed | 2024-10-01T04:04:59Z |
| format | Journal Article |
| id | ntu-10356/161072 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T04:04:59Z |
| publishDate | 2022 |
| record_format | dspace |
| spelling | ntu-10356/1610722022-08-15T01:43:25Z Transfer learning enhanced AR spatial registration for facility maintenance management Chen, Keyu Yang, Jianfei Cheng, Jack C. P. Chen, Weiwei Li, Chun Ting School of Electrical and Electronic Engineering Engineering::Electrical and electronic engineering AR Spatial Registration Facility Maintenance Management Augmented reality (AR), which requires a spatial registration technique, has proved to greatly improve the efficiency of facility maintenance management (FMM) activities. Being one of the most promising techniques for indoor localization, Wi-Fi fingerprinting has been widely used for AR spatial registration. However, localization accuracy of Wi-Fi fingerprinting decreases over time due to dynamics of environmental factors. Readings from different mobile devices can also affect the accuracy negatively. In this paper, a transfer learning technique named transferable CNN-LSTM is proposed for improving the robustness of Wi-Fi fingerprinting while implementing AR in FMM activities. Convolutional neural network (CNN), embedded with long short term memory (LSTM) networks, is utilized to predict the location of unlabeled fingerprints. Multiple kernel variant of maximum mean discrepancy (MK-MMD) is adopted to reduce the distribution difference between the source domain and the target domain, so that the location of the newly collected unlabeled fingerprints can be predicted accurately. As shown in the experimental validation, the transferable CNN-LSTM can achieve an accuracy of 97.1% in short-term (without significant environmental changes) spatial registration, 87.8% in long-term (with significant environmental changes) spatial registration, and around 90% in multi-device spatial registration, indicating a higher accuracy and better robustness over other conventional approaches. 2022-08-15T01:43:25Z 2022-08-15T01:43:25Z 2020 Journal Article Chen, K., Yang, J., Cheng, J. C. P., Chen, W. & Li, C. T. (2020). Transfer learning enhanced AR spatial registration for facility maintenance management. Automation in Construction, 113, 103135-. https://dx.doi.org/10.1016/j.autcon.2020.103135 0926-5805 https://hdl.handle.net/10356/161072 10.1016/j.autcon.2020.103135 2-s2.0-85080088943 113 103135 en Automation in Construction © 2020 Elsevier B.V. All rights reserved. |
| spellingShingle | Engineering::Electrical and electronic engineering AR Spatial Registration Facility Maintenance Management Chen, Keyu Yang, Jianfei Cheng, Jack C. P. Chen, Weiwei Li, Chun Ting Transfer learning enhanced AR spatial registration for facility maintenance management |
| title | Transfer learning enhanced AR spatial registration for facility maintenance management |
| title_full | Transfer learning enhanced AR spatial registration for facility maintenance management |
| title_fullStr | Transfer learning enhanced AR spatial registration for facility maintenance management |
| title_full_unstemmed | Transfer learning enhanced AR spatial registration for facility maintenance management |
| title_short | Transfer learning enhanced AR spatial registration for facility maintenance management |
| title_sort | transfer learning enhanced ar spatial registration for facility maintenance management |
| topic | Engineering::Electrical and electronic engineering AR Spatial Registration Facility Maintenance Management |
| url | https://hdl.handle.net/10356/161072 |
| work_keys_str_mv | AT chenkeyu transferlearningenhancedarspatialregistrationforfacilitymaintenancemanagement AT yangjianfei transferlearningenhancedarspatialregistrationforfacilitymaintenancemanagement AT chengjackcp transferlearningenhancedarspatialregistrationforfacilitymaintenancemanagement AT chenweiwei transferlearningenhancedarspatialregistrationforfacilitymaintenancemanagement AT lichunting transferlearningenhancedarspatialregistrationforfacilitymaintenancemanagement |