Indoor Visible-Light 3D Positioning System Based on GRU Neural Network
With the continuous development of artificial intelligence technology, visible-light positioning (VLP) based on machine learning and deep learning algorithms has become a research hotspot for indoor positioning technology. To improve the accuracy of robot positioning, we established a three-dimensio...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-05-01
|
| Series: | Photonics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2304-6732/10/6/633 |
| _version_ | 1797593044002799616 |
|---|---|
| author | Wuju Yang Ling Qin Xiaoli Hu Desheng Zhao |
| author_facet | Wuju Yang Ling Qin Xiaoli Hu Desheng Zhao |
| author_sort | Wuju Yang |
| collection | DOAJ |
| description | With the continuous development of artificial intelligence technology, visible-light positioning (VLP) based on machine learning and deep learning algorithms has become a research hotspot for indoor positioning technology. To improve the accuracy of robot positioning, we established a three-dimensional (3D) positioning system of visible-light consisting of two LED lights and three photodetectors. In this system, three photodetectors are located on the robot’s head. We considered the impact of line-of-sight (LOS) and non-line-of-sight (NLOS) links on the received signals and used gated recurrent unit (GRU) neural networks to deal with nonlinearity in the system. To address the problem of poor stability during GRU network training, we used a learning rate attenuation strategy to improve the performance of the GRU network. The simulation results showed that the average positioning error of the system was 2.69 cm in a space of 4 m × 4 m × 3 m when only LOS links were considered and 2.66 cm when both LOS and NLOS links were considered with 95% of the positioning errors within 7.88 cm. For two-dimensional (2D) positioning with a fixed positioning height, 80% of the positioning error was within 9.87 cm. This showed that the system had a high anti-interference ability, could achieve centimeter-level positioning accuracy, and met the requirements of robot indoor positioning. |
| first_indexed | 2024-03-11T02:01:54Z |
| format | Article |
| id | doaj.art-5133e1f7bb2f464f9d27b0dcc743bb6d |
| institution | Directory Open Access Journal |
| issn | 2304-6732 |
| language | English |
| last_indexed | 2024-03-11T02:01:54Z |
| publishDate | 2023-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Photonics |
| spelling | doaj.art-5133e1f7bb2f464f9d27b0dcc743bb6d2023-11-18T12:07:33ZengMDPI AGPhotonics2304-67322023-05-0110663310.3390/photonics10060633Indoor Visible-Light 3D Positioning System Based on GRU Neural NetworkWuju Yang0Ling Qin1Xiaoli Hu2Desheng Zhao3College of Information Engineering, Inner Mongolia University of Science and Technology, Baotou 014010, ChinaCollege of Information Engineering, Inner Mongolia University of Science and Technology, Baotou 014010, ChinaCollege of Information Engineering, Inner Mongolia University of Science and Technology, Baotou 014010, ChinaCollege of Information Engineering, Inner Mongolia University of Science and Technology, Baotou 014010, ChinaWith the continuous development of artificial intelligence technology, visible-light positioning (VLP) based on machine learning and deep learning algorithms has become a research hotspot for indoor positioning technology. To improve the accuracy of robot positioning, we established a three-dimensional (3D) positioning system of visible-light consisting of two LED lights and three photodetectors. In this system, three photodetectors are located on the robot’s head. We considered the impact of line-of-sight (LOS) and non-line-of-sight (NLOS) links on the received signals and used gated recurrent unit (GRU) neural networks to deal with nonlinearity in the system. To address the problem of poor stability during GRU network training, we used a learning rate attenuation strategy to improve the performance of the GRU network. The simulation results showed that the average positioning error of the system was 2.69 cm in a space of 4 m × 4 m × 3 m when only LOS links were considered and 2.66 cm when both LOS and NLOS links were considered with 95% of the positioning errors within 7.88 cm. For two-dimensional (2D) positioning with a fixed positioning height, 80% of the positioning error was within 9.87 cm. This showed that the system had a high anti-interference ability, could achieve centimeter-level positioning accuracy, and met the requirements of robot indoor positioning.https://www.mdpi.com/2304-6732/10/6/633robotvisible-light positioning (VLP)three-dimensional (3D)line-of-sight (LOS) and non-line-of-sight (NLOS) linksgated recurrent units (GRU) neural networkslearning rate decay strategy |
| spellingShingle | Wuju Yang Ling Qin Xiaoli Hu Desheng Zhao Indoor Visible-Light 3D Positioning System Based on GRU Neural Network Photonics robot visible-light positioning (VLP) three-dimensional (3D) line-of-sight (LOS) and non-line-of-sight (NLOS) links gated recurrent units (GRU) neural networks learning rate decay strategy |
| title | Indoor Visible-Light 3D Positioning System Based on GRU Neural Network |
| title_full | Indoor Visible-Light 3D Positioning System Based on GRU Neural Network |
| title_fullStr | Indoor Visible-Light 3D Positioning System Based on GRU Neural Network |
| title_full_unstemmed | Indoor Visible-Light 3D Positioning System Based on GRU Neural Network |
| title_short | Indoor Visible-Light 3D Positioning System Based on GRU Neural Network |
| title_sort | indoor visible light 3d positioning system based on gru neural network |
| topic | robot visible-light positioning (VLP) three-dimensional (3D) line-of-sight (LOS) and non-line-of-sight (NLOS) links gated recurrent units (GRU) neural networks learning rate decay strategy |
| url | https://www.mdpi.com/2304-6732/10/6/633 |
| work_keys_str_mv | AT wujuyang indoorvisiblelight3dpositioningsystembasedongruneuralnetwork AT lingqin indoorvisiblelight3dpositioningsystembasedongruneuralnetwork AT xiaolihu indoorvisiblelight3dpositioningsystembasedongruneuralnetwork AT deshengzhao indoorvisiblelight3dpositioningsystembasedongruneuralnetwork |