Abstract: An artificial intelligence enhanced star identification algorithm is proposed for star trackers in lost-in-space mode. A convolutional neural network model based on Vgg16 is used in the artificial intelligence algorithm to classify star images. The training dataset is constructed to achieve the networks’ optimal performance. Simulation results show that the proposed algorithm is highly robust to many kinds of noise, including position noise, magnitude noise, false stars, and the tracker’s angular velocity. With a deep convolutional neural network, the identification accuracy is maintained at 96% despite noise and interruptions, which is a significant improvement to traditional pyramid and grid algorithms.

一种基于人工智能的星图识别算法

[1]Aghaei M, Moghaddam HA, 2016. Grid star identification improvement using optimization approaches. IEEE Trans Aerosp Electron Syst, 52(5):2080-2090.

[2]Cole CL, Crassidis JL, 2006. Fast star-pattern recognition using planar triangles. J Guid Contr Dynam, 29(1):64-71.

[3]Cheng J, Wang PS, Li G, 2018. Recent advances in efficient computation of deep convolutional neural networks. Front Inform Technol Electron Eng, 19(1):64-77.

[4]Hancock BR, Stirbl RC, Cunningham TJ, et al., 2020. CMOS active pixel sensor specific performance effects on star tracker/imager position accuracy. Symp on Integrated Optics, p.43-53.

[5]Hernández EA, Alonso MA, Chávez E, et al. 2017. Robust polygon recognition method with similarity invariants applied to star identification. Adv Space Res, 59(4):1095-1111.

[6]Hong J, Dickerson JA, 2000. Neural-network-based autonomous star identification algorithm. J Guid Contr Dynam, 23(4):728-735.

[7]Jia YQ, Shelhamer E, Donahue J, et al., 2014. Caffe: convolutional architecture for fast feature embedding. Proc 22^nd ACM Int Conf on Multimedia, p.675-678.

[8]Jing Y, Liang W, 2012. An improved star identification method based on neural network. Proc IEEE 10^th Int Conf on Industrial Informatics, p.118-123.

[9]Juang JN, Kim HY, Junkins JL, 2003. An efficient and robust singular value method for star pattern recognition and attitude determination. NASA/TM-2003-212142, NASA Langley Research Center, Hampton, USA.

[10]Kim K, Bang H, 2020. Algorithm with patterned singular value approach for highly reliable autonomous star identification. Sensors, 20(2):374.

[11]Krizhevsky A, Sutskever I, Hinton GE, 2017. ImageNet classification with deep convolutional neural networks. Commun ACM, 60(6):94-90.

[12]Kruzhilov IS, 2012. Evaluation of instrument stellar magnitudes without recourse to data as to star spectral classes. J Appl Remote Sens, 6(1):063537.

[13]Kumar M, Mortari D, Junkins JL, 2010. An analytical approach to star identification reliability. Acta Astronaut, 66(3-4):508-515.

[14]LeCun Y, Bengio Y, Hinton G, 2015. Deep learning. Nature, 521(7553):436-444.

[15]Liebe CC, 1992. Pattern recognition of star constellations for spacecraft applications. IEEE Aerosp Electron Syst Mag, 7(6):34-41.

[16]Liu B, Dai Y, Li X, et al., 2003. Building text classifiers using positive and unlabeled examples. Proc 3^rd IEEE Int Conf on Data Mining, p.179-188.

[17]Mehta DS, Chen SS, Low KS, 2018. A robust star identification algorithm with star shortlisting. Adv Space Res, 61(10):2647-2660.

[18]Mortari D, Samaan MA, Bruccoleri C, 2004. The pyramid star identification technique. Navigation, 51(3):171-183.

[19]Na M, Zheng DN, Jia PF, 2009. Modified grid algorithm for noisy all-sky autonomous star identification. IEEE Trans Aerosp Electron Syst, 45(2):516-522.

[20]Opromolla R, Fasano G, Rufino G, et al., 2017. A new star tracker concept for satellite attitude determination based on a multi-purpose panoramic camera. Acta Astronaut, 140:166-175.

[21]Padgett C, Kreutz-Delgado K, 1997. A grid algorithm for autonomous star identification. IEEE Trans Aerosp Electron Syst, 33(1):202-213.

[22]Qiu JT, Wang J, Yao S, et al., 2016. Going deeper with embedded FPGA platform for convolutional neural network. Proc ACM/SIGDA Int Symp on Field-Programmable Gate Arrays, p.26-35.

[23]Quan W, Fang JC, 2010. A star recognition method based on the adaptive ant colony algorithm for star sensors. Sensors, 10(3):1955-1966.

[24]Roberts P, Walker R, 2005. Application of a counter propagation neural network for star identification. AIAA Guidance, Navigation, and Control Conf and Exhibit, p.15-18.

[25]Roshanian J, Yazdani S, Ebrahimi M, 2016. Star identification based on Euclidean distance transform, Voronoi tessellation, and k-nearest neighbor classification. IEEE Trans Aerosp Electron Syst, 52(6):2940-2949.

[26]Samirbhai MD, Chen S, Low KS, 2019. A hamming distance and spearman correlation based star identification algorithm. IEEE Trans Aerosp Electron Syst, 55(1):17-30.

[27]Schiattarella V, Spiller D, Curti F, 2017. A novel star identification technique robust to high presence of false objects: the multi-poles algorithm. Adv Space Res, 59(8):2133-2147.

[28]Simonyan K, Zisserman A, 2015. Very deep convolutional networks for large-scale image recognition. https://arxiv.org/abs/1409.1556

[29]Spratling BB IV, Mortari D, 2009. A survey on star identification algorithms. Algorithms, 2(1):93-107.

[30]Sun ZP, Qi XD, Jin G, et al., 2017. Ellipticity pivot star method for autonomous star identification. Optik, 137:1-5.

[31]Szegedy C, Liu W, Jia YQ, et al., 2015. Going deeper with convolutions. Proc IEEE Conf on Computer Vision and Pattern Recognition, p.1-9.

[32]Wang XC, Sun CH, Sun T, 2019. A novel 2-step validation algorithm for lost-in-space star identification. IEEE Trans Aerosp Electron Syst, 56(3):2272-2279.

[33]Wei X, Wen DS, Song ZX, et al., 2019. A star identification algorithm based on radial and dynamic cyclic features of star pattern. Adv Space Res, 63(7):2245-2259.

[34]Zhang GJ, 2017. Star identification: methods, techniques, and algorithms. Springer, Berlin, Germany.

[35]Zhang GJ, Wei XG, Jiang J, 2008. Full-sky autonomous star identification based on radial and cyclic features of star pattern. Image Vis Comput, 26(7):891-897.

[36]Zhang QS, Zhu SC, 2018. Visual interpretability for deep learning: a survey. Front Inform Technol Electron Eng, 19(1):27-39.