An Integration of Deep Neural Network-Based Extended Kalman Filter (DNN-EKF) Method in Ultra-Wideband (UWB) Localization for Distance Loss Optimization
This paper examines the critical role of indoor positioning for robots, with a particular focus on small and confined spaces such as homes, warehouses, and similar environments. We develop an algorithm by integrating deep neural networks (DNNs) with the extended Kalman filter (EKF) method, which is...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-11-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/24/23/7643 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850106760576106496 |
|---|---|
| author | Chanthol Eang Seungjae Lee |
| author_facet | Chanthol Eang Seungjae Lee |
| author_sort | Chanthol Eang |
| collection | DOAJ |
| description | This paper examines the critical role of indoor positioning for robots, with a particular focus on small and confined spaces such as homes, warehouses, and similar environments. We develop an algorithm by integrating deep neural networks (DNNs) with the extended Kalman filter (EKF) method, which is known as DNN-EKF, to obtain an accurate indoor localization for ensuring precise and reliable robot movements within the use of Ultra-Wideband (UWB) technology. The study introduces a novel methodology that combines advanced technology, including DNN, filtering techniques, specifically the EKF and UWB technology, with the objective of enhancing the accuracy of indoor localization systems. The objective of integrating these technologies is to develop a more robust and dependable solution for robot navigation in challenging indoor environments. The proposed approach combines a DNN with the EKF to significantly improve indoor localization accuracy for mobile robots. The results clearly show that the proposed model outperforms existing methods, including NN-EKF, LPF-EKF, and other traditional approaches. In particular, the DNN-EKF method achieves optimal performance with the least distance loss compared to NN-EKF and LPF-EKF. These results highlight the superior effectiveness of the DNN-EKF method in providing precise localization in indoor environments, especially when utilizing UWB technology. This makes the model highly suitable for real-time robotic applications, particularly in dynamic and noisy environments. |
| format | Article |
| id | doaj-art-e312120b3c98412087e9f861f0de75e8 |
| institution | OA Journals |
| issn | 1424-8220 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-e312120b3c98412087e9f861f0de75e82025-08-20T02:38:45ZengMDPI AGSensors1424-82202024-11-012423764310.3390/s24237643An Integration of Deep Neural Network-Based Extended Kalman Filter (DNN-EKF) Method in Ultra-Wideband (UWB) Localization for Distance Loss OptimizationChanthol Eang0Seungjae Lee1Department of Computer Science and Engineering, Intelligent Robot Research Institute, Sun Moon University, Asan 31460, Republic of KoreaDepartment of Computer Science and Engineering, Intelligent Robot Research Institute, Sun Moon University, Asan 31460, Republic of KoreaThis paper examines the critical role of indoor positioning for robots, with a particular focus on small and confined spaces such as homes, warehouses, and similar environments. We develop an algorithm by integrating deep neural networks (DNNs) with the extended Kalman filter (EKF) method, which is known as DNN-EKF, to obtain an accurate indoor localization for ensuring precise and reliable robot movements within the use of Ultra-Wideband (UWB) technology. The study introduces a novel methodology that combines advanced technology, including DNN, filtering techniques, specifically the EKF and UWB technology, with the objective of enhancing the accuracy of indoor localization systems. The objective of integrating these technologies is to develop a more robust and dependable solution for robot navigation in challenging indoor environments. The proposed approach combines a DNN with the EKF to significantly improve indoor localization accuracy for mobile robots. The results clearly show that the proposed model outperforms existing methods, including NN-EKF, LPF-EKF, and other traditional approaches. In particular, the DNN-EKF method achieves optimal performance with the least distance loss compared to NN-EKF and LPF-EKF. These results highlight the superior effectiveness of the DNN-EKF method in providing precise localization in indoor environments, especially when utilizing UWB technology. This makes the model highly suitable for real-time robotic applications, particularly in dynamic and noisy environments.https://www.mdpi.com/1424-8220/24/23/7643deep neural networks (DNNs)extended Kalman filter (EKF)ultra-wideband (UWB) technologydeep neural network-based extended Kalman filter (DNN-EKF) |
| spellingShingle | Chanthol Eang Seungjae Lee An Integration of Deep Neural Network-Based Extended Kalman Filter (DNN-EKF) Method in Ultra-Wideband (UWB) Localization for Distance Loss Optimization Sensors deep neural networks (DNNs) extended Kalman filter (EKF) ultra-wideband (UWB) technology deep neural network-based extended Kalman filter (DNN-EKF) |
| title | An Integration of Deep Neural Network-Based Extended Kalman Filter (DNN-EKF) Method in Ultra-Wideband (UWB) Localization for Distance Loss Optimization |
| title_full | An Integration of Deep Neural Network-Based Extended Kalman Filter (DNN-EKF) Method in Ultra-Wideband (UWB) Localization for Distance Loss Optimization |
| title_fullStr | An Integration of Deep Neural Network-Based Extended Kalman Filter (DNN-EKF) Method in Ultra-Wideband (UWB) Localization for Distance Loss Optimization |
| title_full_unstemmed | An Integration of Deep Neural Network-Based Extended Kalman Filter (DNN-EKF) Method in Ultra-Wideband (UWB) Localization for Distance Loss Optimization |
| title_short | An Integration of Deep Neural Network-Based Extended Kalman Filter (DNN-EKF) Method in Ultra-Wideband (UWB) Localization for Distance Loss Optimization |
| title_sort | integration of deep neural network based extended kalman filter dnn ekf method in ultra wideband uwb localization for distance loss optimization |
| topic | deep neural networks (DNNs) extended Kalman filter (EKF) ultra-wideband (UWB) technology deep neural network-based extended Kalman filter (DNN-EKF) |
| url | https://www.mdpi.com/1424-8220/24/23/7643 |
| work_keys_str_mv | AT chantholeang anintegrationofdeepneuralnetworkbasedextendedkalmanfilterdnnekfmethodinultrawidebanduwblocalizationfordistancelossoptimization AT seungjaelee anintegrationofdeepneuralnetworkbasedextendedkalmanfilterdnnekfmethodinultrawidebanduwblocalizationfordistancelossoptimization AT chantholeang integrationofdeepneuralnetworkbasedextendedkalmanfilterdnnekfmethodinultrawidebanduwblocalizationfordistancelossoptimization AT seungjaelee integrationofdeepneuralnetworkbasedextendedkalmanfilterdnnekfmethodinultrawidebanduwblocalizationfordistancelossoptimization |