Abstract
Benefiting from multi-constellation Global Navigation Satellite Systems (GNSS), more and more visible satellites can be used to improve user positioning performance. However, due to limited tracking receiver channels and power consumption, and other issues, it may be not possible, or desirable, to use all satellites in view for positioning. The optimal subset is generally selected from all possible satellite combinations to minimize either Geometric Dilution of Precision (GDOP) or weighted GDOP (WGDOP). However, this brute force approach is difficult to implement in real-time applications due to the time- and power-consuming calculation of the DOP values. As an alternative to a brute force satellite selection procedure, the authors propose an end-to-end deep learning network for satellite selection based on the PointNet and VoxelNet networks. The satellite selection is converted to a satellite segmentation problem, with specified input channel for each satellite and two class labels, one for selected satellites and the other for those not selected. The aim of the satellite segmentation is that a fixed number of satellites with the minimum GDOP/WGDOP value can be segmented from any feeding order of input satellites. To validate the proposed satellite segmentation network, training and test data from 220 IGS stations tracking GPS and GLONASS satellites were used. The segmentation performance using different architectures and representations of input channels, including receiver-to-satellite unit vector and elevation and azimuth, were compared. It was found that the input channel with elevation and azimuth can achieve better performance than using the receiver-to-satellite unit vector, and an architecture with stacked feature encoding (FE) layers has better satellite segmentation performance than one without stacked FE layers. In addition, the models with GDOP and WGDOP criteria for selecting 9 and 12 satellites were trained. It was demonstrated that the satellite segmentation network was about 90 times faster than using the brute force approach. Furthermore, all the trained models can effectively select the satellites making the most contribution to the desired GDOP/WGDOP value. Approximately 99% of the tests had GDOP and WGDOP value differences smaller than 0.03 and 0.2, respectively, between the predicted subset and the optimal subset.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.References
Abadi M et al (2016) Tensorflow: large-scale machine learning on heterogeneous distributed systems (arXiv preprint: arXiv:1603.04467)
Azami H, Sanei S (2014) GPS GDOP classification via improved neural network trainings and principal component analysis. Int J Electron 101(9):1300–1313
Azami H, Mosavi MR, Sanei S (2013) Classification of GPS satellites using improved back propagation training algorithms. Wirel Pers Commun 71(2):789–803
Blanco-Delgado N, Nunes FD (2010) Satellite selection method for multi-constellation GNSS using convex geometry. IEEE Trans Veh Technol 59(9):4289–4297
Blanco-Delgado N, Nunes FD, Seco-Granados G (2017) On the relation between GDOP and the volume described by the user-to-satellite unit vectors for GNSS positioning. GPS Solut 21(3):1139–1147
Dauphin Y, Pascanu R, Gulcehre C, Cho K, Ganguli S, Bengio Y (2014) Identifying and attacking the saddle point problem in high-dimensional non-convex optimization. In: Advances in neural information processing systems, Montreal, 08–13 Dec 2014, vol 2 pp 2933–2941
De Boer PT, Kroese DP, Mannor S, Rubinstein RY (2005) A tutorial on the cross-entropy method. Ann Oper Res 134(1):19–67
Doong SH (2009) A closed-form formula for GPS GDOP computation. GPS Solut 13(3):183–190
Grilli E, Menna F, Remondino F (2017) A review of point clouds segmentation and classification algorithms. Int Arch Photogramm Remote Sens Spat Inf Sci 42(2):W3
Jwo DJ, Lai CC (2007) Neural network-based GPS GDOP approximation and classification. GPS Solut 11(1):51–60
Kingma DP, Ba J (2014) Adam: a method for stochastic optimization (arXiv preprint: arXiv:1412.6980)
Kong J, Mao X, Li S (2014) BDS/GPS satellite selection algorithm based on polyhedron volumetric method. In: 2014 IEEE/SICE international symposium on system integration, Tokyo, 13–15 Dec 2014, pp 340–345
Li G, Xu C, Zhang P, Hu C (2012) A modified satellite selection algorithm based on satellite contribution for GDOP in GNSS. In: Advances in mechanical and electronic engineering, volume 1, pp 415–421
Li G, Wu J, Liu W, Zhao C (2016) A new approach of satellite selection for multi-constellation integrated navigation system. In: China satellite navigation conference (CSNC) 2016 proceedings, volume III, pp 359–371
Liu M, Fortin M-A, Landry R (2009) A recursive quasi-optimal fast satellite selection method for GNSS receivers. In: Proceedings of the ION GNSS 2009, Institute of Navigation, Savannah, 22–25 September 2009, pp 2061–2071
Mosavi M (2011) Applying genetic algorithm to fast and precise selection of GPS satellites. Asian J Appl Sci Eng 4(3):229–237
Peng A, Ou G, Li G (2014) Fast satellite selection method for multi-constellation Global Navigation Satellite System under obstacle environments. IET Radar Sonar Navig 8(9):1051–1058
Qi CR, Su H, Mo K, Guibas LJ (2016) PointNet: deep learning on point sets for 3D classification and segmentation (arXiv preprint arXiv:161200593)
Roongpiboonsopit D, Karimi HA (2009) A multi-constellations satellite selection algorithm for integrated global navigation satellite systems. J Intell Transp Syst 13(3):127–141
Simon D, El-Sherief H (1995) Navigation satellite selection using neural networks. Neurocomputing 7(3):247–258
Swaszek PF, Hartnett RJ, Seals KC, Swaszek R (2017) A temporal algorithm for satellite subset selection in multi-constellation GNSS. In: Proceedings of the ION ITM 2017, Institute of Navigation, Monterey, 30 Jan–2 Feb 2017, pp 1147–1159
Teng Y, Wang J (2016) A closed-form formula to calculate geometric dilution of precision (GDOP) for multi-GNSS constellations. GPS Solut 20(3):331–339
Walter T, Blanch J, Kropp V (2016) Satellite selection for multi-constellation SBAS. In: Proceedings of the ION GNSS 2016, Institute of Navigation, Portland, 12–16 September 2016, pp 1350–1359
Wei M, Wang J, Li J (2012) A new satellite selection algorithm for real-time application. In: 2012 international conference on systems and informatics (ICSAI2012), Yantai, pp 2567–2570
Wu CH, Su WH, Ho YW (2011) A study on GPS GDOP approximation using support-vector machines. IEEE Trans Instrum Meas 60(1):137–145
Zarei N (2014) Artificial intelligence approaches for GPS GDOP classification. Int J Comput Appl 96(16):16–21
Zhang M, Zhang J (2009) A fast satellite selection algorithm: beyond four satellites. IEEE J Sel Top Signal Process 3(5):740–747
Zhou Y, Tuzel O (2017) VoxelNet: end-to-end learning for point cloud based 3D object detection (arXiv preprint: arXiv:171106396)
Zhu S (2018) An optimal satellite selection model of global navigation satellite system based on genetic algorithm. In: China satellite navigation conference (CSNC) 2018 proceedings
Acknowledgements
This research is supported by the Chinese Scholarship Council (CSC) awarded to the Panpan Huang.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Huang, P., Rizos, C. & Roberts, C. Satellite selection with an end-to-end deep learning network. GPS Solut 22, 108 (2018). https://doi.org/10.1007/s10291-018-0776-0
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10291-018-0776-0