SmartFPS: Neural network based wireless-inertial fusion positioning system
Current wireless-inertial fusion positioning systems widely adopt empirical propagation models of wireless signals and filtering algorithms such as the Kalman filter or the particle filter. However, empirical models of system and noise usually have lower accuracy in a practical positioning scenario....
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-02-01
|
| Series: | Frontiers in Neurorobotics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnbot.2023.1121623/full |
| _version_ | 1811167227622195200 |
|---|---|
| author | Luchi Hua Yuan Zhuang Yuan Zhuang Yuan Zhuang Jun Yang |
| author_facet | Luchi Hua Yuan Zhuang Yuan Zhuang Yuan Zhuang Jun Yang |
| author_sort | Luchi Hua |
| collection | DOAJ |
| description | Current wireless-inertial fusion positioning systems widely adopt empirical propagation models of wireless signals and filtering algorithms such as the Kalman filter or the particle filter. However, empirical models of system and noise usually have lower accuracy in a practical positioning scenario. The biases of predetermined parameters would enlarge the positioning error through layers of systems. Instead of dealing with empirical models, this paper proposes a fusion positioning system based on an end-to-end neural network, along with a transfer learning strategy for improving the performance of neural network models for samples with different distributions. Verified by Bluetooth-inertial positioning in a whole floor scenario, the mean positioning error of the fusion network was 0.506 m. The proposed transfer learning method improved the accuracy of the step length and rotation angle of different pedestrians by 53.3%, the Bluetooth positioning accuracy of various devices by 33.4%, and the average positioning error of the fusion system by 31.6%. The results showed that our proposed methods outperformed filter-based methods in challenging indoor environments. |
| first_indexed | 2024-04-10T16:05:42Z |
| format | Article |
| id | doaj.art-39ce813f33e64f839f6dd7cfdcbf2011 |
| institution | Directory Open Access Journal |
| issn | 1662-5218 |
| language | English |
| last_indexed | 2024-04-10T16:05:42Z |
| publishDate | 2023-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neurorobotics |
| spelling | doaj.art-39ce813f33e64f839f6dd7cfdcbf20112023-02-10T05:02:39ZengFrontiers Media S.A.Frontiers in Neurorobotics1662-52182023-02-011710.3389/fnbot.2023.11216231121623SmartFPS: Neural network based wireless-inertial fusion positioning systemLuchi Hua0Yuan Zhuang1Yuan Zhuang2Yuan Zhuang3Jun Yang4National Application Specific Integrated Circuit Center, Southeast University, Nanjing, ChinaState Key Laboratory of Information Engineering in Surveying, Mapping, and Remote Sensing, Wuhan University, Wuhan, ChinaHubei Luojia Laboratory, Wuhan, ChinaWuhan University Shenzhen Research Institute, Shenzhen, ChinaNational Application Specific Integrated Circuit Center, Southeast University, Nanjing, ChinaCurrent wireless-inertial fusion positioning systems widely adopt empirical propagation models of wireless signals and filtering algorithms such as the Kalman filter or the particle filter. However, empirical models of system and noise usually have lower accuracy in a practical positioning scenario. The biases of predetermined parameters would enlarge the positioning error through layers of systems. Instead of dealing with empirical models, this paper proposes a fusion positioning system based on an end-to-end neural network, along with a transfer learning strategy for improving the performance of neural network models for samples with different distributions. Verified by Bluetooth-inertial positioning in a whole floor scenario, the mean positioning error of the fusion network was 0.506 m. The proposed transfer learning method improved the accuracy of the step length and rotation angle of different pedestrians by 53.3%, the Bluetooth positioning accuracy of various devices by 33.4%, and the average positioning error of the fusion system by 31.6%. The results showed that our proposed methods outperformed filter-based methods in challenging indoor environments.https://www.frontiersin.org/articles/10.3389/fnbot.2023.1121623/fullindoor positioningwireless positioningKalman filterdeep learningtransfer learning |
| spellingShingle | Luchi Hua Yuan Zhuang Yuan Zhuang Yuan Zhuang Jun Yang SmartFPS: Neural network based wireless-inertial fusion positioning system Frontiers in Neurorobotics indoor positioning wireless positioning Kalman filter deep learning transfer learning |
| title | SmartFPS: Neural network based wireless-inertial fusion positioning system |
| title_full | SmartFPS: Neural network based wireless-inertial fusion positioning system |
| title_fullStr | SmartFPS: Neural network based wireless-inertial fusion positioning system |
| title_full_unstemmed | SmartFPS: Neural network based wireless-inertial fusion positioning system |
| title_short | SmartFPS: Neural network based wireless-inertial fusion positioning system |
| title_sort | smartfps neural network based wireless inertial fusion positioning system |
| topic | indoor positioning wireless positioning Kalman filter deep learning transfer learning |
| url | https://www.frontiersin.org/articles/10.3389/fnbot.2023.1121623/full |
| work_keys_str_mv | AT luchihua smartfpsneuralnetworkbasedwirelessinertialfusionpositioningsystem AT yuanzhuang smartfpsneuralnetworkbasedwirelessinertialfusionpositioningsystem AT yuanzhuang smartfpsneuralnetworkbasedwirelessinertialfusionpositioningsystem AT yuanzhuang smartfpsneuralnetworkbasedwirelessinertialfusionpositioningsystem AT junyang smartfpsneuralnetworkbasedwirelessinertialfusionpositioningsystem |