Download PDFOpen PDF in browser

Airborne Navigation By Geomagnetic Field Based on LSTM

9 pagesPublished: September 20, 2022

Abstract

To solve the problem of low navigation accuracy of traditional geomagnetic matching navigation algorithm, a geomagnetic navigation and positioning algorithm based on long-term and short-term memory neural network (LSTM) is proposed in this paper. In this algorithm, the corresponding geodetic coordinates are derived from geomagnetic measurements based on least-square linear fitting. Therefore, the geomagnetic matching is implemented. Then the position of the aircraft at the next time is predicted by the LSTM algorithm. Furthermore, the corresponding geodetic coordinates derived from the geomagnetic sensor are modified to complete the geomagnetic navigation and positioning. In this paper, the geomagnetic field data of a certain region are obtained by IGRF-13. The multi-time simulated flight trajectory is used for simulation experiments. The results show that the proposed methods are reliable to transform the geomagnetic measurements to geodetic coordinates. Also, the artificial intelligent method is to make up for the measurement error of the geomagnetic sensor and improve the accuracy of the geomagnetic navigation.

Keyphrases: fitting, geomagnetic navigation, least square, lstm, trajectory prediction

In: Tokuro Matsuo (editor). Proceedings of 11th International Congress on Advanced Applied Informatics, vol 81, pages 132-140.

BibTeX entry
@inproceedings{IIAIAAI2021-Winter:Airborne_Navigation_Geomagnetic_Field,
  author    = {Xin Peng and Yibo Wei and Weibao Zou},
  title     = {Airborne Navigation By Geomagnetic Field Based on LSTM},
  booktitle = {Proceedings of 11th International Congress on Advanced Applied Informatics},
  editor    = {Tokuro Matsuo},
  series    = {EPiC Series in Computing},
  volume    = {81},
  publisher = {EasyChair},
  bibsource = {EasyChair, https://easychair.org},
  issn      = {2398-7340},
  url       = {/publications/paper/4F1g},
  doi       = {10.29007/smqf},
  pages     = {132-140},
  year      = {2022}}
Download PDFOpen PDF in browser