Improving GNSS Positioning Using Neural-Network-Based Corrections
Deep neural networks (DNNs) are a promising tool for global navigation satellite system (GNSS) positioning in the presence of multipath and non-line-of-sight errors, owing to their ability to model complex errors using data. However, developing a DNN for GNSS positioning presents various challenges,...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Institute of Navigation
2022-11-01
|
| Series: | Navigation |
| Online Access: | https://navi.ion.org/content/69/4/navi.548 |
| _version_ | 1827586413207486464 |
|---|---|
| author | Ashwin V. Kanhere Shubh Gupta Akshay Shetty Grace Gao |
| author_facet | Ashwin V. Kanhere Shubh Gupta Akshay Shetty Grace Gao |
| author_sort | Ashwin V. Kanhere |
| collection | DOAJ |
| description | Deep neural networks (DNNs) are a promising tool for global navigation satellite system (GNSS) positioning in the presence of multipath and non-line-of-sight errors, owing to their ability to model complex errors using data. However, developing a DNN for GNSS positioning presents various challenges, such as (a) poor numerical conditioning caused by large variations in measurements and position values across the globe, (b) varying number and order within the set of measurements due to changing satellite visibility, and (c) overfitting to available data. In this work, we address the aforementioned challenges and propose an approach for GNSS positioning by applying DNN-based corrections to an initial position guess. Our DNN learns to output the position correction using the set of pseudorange residuals and satellite line-of-sight vectors as inputs. The limited variation in these input and output values improves the numerical conditioning for our DNN. We design our DNN architecture to combine information from the available GNSS measurements, which vary both in number and order, by leveraging recent advancements in set-based deep learning methods. Furthermore, we present a data augmentation strategy to reduce overfitting in the DNN by randomizing the initial position guesses. We, first, perform simulations and show an improvement in the initial positioning error when our DNN-based corrections are applied. After this, we demonstrate that our approach outperforms a weighted least squares (WLS) baseline on real-world data. Our implementation is available at github.com/Stanford-NavLab/deep_gnss. |
| first_indexed | 2024-03-09T00:00:36Z |
| format | Article |
| id | doaj.art-54d42d72ef134ed48c28df89f0d72ea0 |
| institution | Directory Open Access Journal |
| issn | 2161-4296 |
| language | English |
| last_indexed | 2024-03-09T00:00:36Z |
| publishDate | 2022-11-01 |
| publisher | Institute of Navigation |
| record_format | Article |
| series | Navigation |
| spelling | doaj.art-54d42d72ef134ed48c28df89f0d72ea02023-12-12T17:44:12ZengInstitute of NavigationNavigation2161-42962022-11-0169410.33012/navi.548navi.548Improving GNSS Positioning Using Neural-Network-Based CorrectionsAshwin V. KanhereShubh GuptaAkshay ShettyGrace GaoDeep neural networks (DNNs) are a promising tool for global navigation satellite system (GNSS) positioning in the presence of multipath and non-line-of-sight errors, owing to their ability to model complex errors using data. However, developing a DNN for GNSS positioning presents various challenges, such as (a) poor numerical conditioning caused by large variations in measurements and position values across the globe, (b) varying number and order within the set of measurements due to changing satellite visibility, and (c) overfitting to available data. In this work, we address the aforementioned challenges and propose an approach for GNSS positioning by applying DNN-based corrections to an initial position guess. Our DNN learns to output the position correction using the set of pseudorange residuals and satellite line-of-sight vectors as inputs. The limited variation in these input and output values improves the numerical conditioning for our DNN. We design our DNN architecture to combine information from the available GNSS measurements, which vary both in number and order, by leveraging recent advancements in set-based deep learning methods. Furthermore, we present a data augmentation strategy to reduce overfitting in the DNN by randomizing the initial position guesses. We, first, perform simulations and show an improvement in the initial positioning error when our DNN-based corrections are applied. After this, we demonstrate that our approach outperforms a weighted least squares (WLS) baseline on real-world data. Our implementation is available at github.com/Stanford-NavLab/deep_gnss.https://navi.ion.org/content/69/4/navi.548 |
| spellingShingle | Ashwin V. Kanhere Shubh Gupta Akshay Shetty Grace Gao Improving GNSS Positioning Using Neural-Network-Based Corrections Navigation |
| title | Improving GNSS Positioning Using Neural-Network-Based Corrections |
| title_full | Improving GNSS Positioning Using Neural-Network-Based Corrections |
| title_fullStr | Improving GNSS Positioning Using Neural-Network-Based Corrections |
| title_full_unstemmed | Improving GNSS Positioning Using Neural-Network-Based Corrections |
| title_short | Improving GNSS Positioning Using Neural-Network-Based Corrections |
| title_sort | improving gnss positioning using neural network based corrections |
| url | https://navi.ion.org/content/69/4/navi.548 |
| work_keys_str_mv | AT ashwinvkanhere improvinggnsspositioningusingneuralnetworkbasedcorrections AT shubhgupta improvinggnsspositioningusingneuralnetworkbasedcorrections AT akshayshetty improvinggnsspositioningusingneuralnetworkbasedcorrections AT gracegao improvinggnsspositioningusingneuralnetworkbasedcorrections |