The identification accuracy of inertia tensor of combined spacecraft, which is composed by a servicing spacecraft and a captured target, could be easily affected by the measurement noise of angular rate. Due to frequently changing operating environments of combined spacecraft in space, the measurement noise of angular rate can be very complex. In this paper, an inertia tensor identification approach based on deep learning method is proposed to improve the ability of identifying inertia tensor of combined spacecraft in the presence of complex measurement noise. A deep neural network model for identification is constructed and trained by enough training data and a designed learning strategy. To verify the identification performance of the proposed deep neural network model, two testing set with different ranks of measure noises are used for simulation tests. Comparison tests are also delivered among the proposed deep neural network model, recursive least squares identification method, and tradition deep neural network model. The comparison results show that the proposed deep neural network model yields a more accurate and stable identification performance for inertia tensor of combined spacecraft in changeable and complex operating environments.
Flores-AbadAMaOPhamK, et al.A review of space robotics technologies for on-orbit servicing. Progr Aerosp Sci2014; 68: 1–26.
2.
LiQDengZZhangK, et al.Precise attitude control of multirotary-joint solar-power satellite. J Guid Control Dyn2018; 41: 1435–1442.
3.
YangCHouXWangL. Thermal design, analysis and comparison on three concepts of space solar power satellite. Acta Astronaut2017; 137: 382–402.
4.
WeiCZhangYWangH, et al.Inertia parameter identification of space floating target during robotic exploratory grasping. Proc IMechE, Part G: J Aerospace Engineering 2019; 233(11): 4247–4260.
5.
BiondiGMauroSMohtarT, et al.Attitude recovery from feature tracking for estimating angular rate of non-cooperative spacecraft. Mech Syst Signal Process2017; 83: 321–336.
6.
Wilson E, Sutter DW and Mah RW. MCRLS for on-line spacecraft mass- and thruster-property identification. In: Proceedings of the IASTED international conference on intelligent systems and control, Honolulu, HI, 23–25 August 2004.
7.
Wilson E, Sutter DW and Mah RW. Motion-based mass- and thruster-property identification for thruster-controlled spacecraft. In: Proceedings of the 2005 AIAA Infotech@Aerospace conference, Arlington, VA, 26–29 September 2005.
8.
XuWHuZZhangY, et al.On-orbit identifying the inertia parameters of space robotic systems using simple equivalent dynamics. Acta Astronaut2017; 132: 131–142.
9.
Nguyen-HuynhTCSharfI. Adaptive reactionless motion and parameter identification in postcapture of space debris. J Guid Control Dyn2013; 36: 404–414.
10.
ChuZMaYHouY, et al.Inertial parameter identification using contact force information for an unknown object captured by a space manipulator. Acta Astronaut2017; 131: 69–82.
11.
NiZLiuJWuZ, et al.Identification of the state-space model and payload mass parameter of a flexible space manipulator using a recursive subspace tracking method. Chin J Aeronaut2018; 32: 513–530.
12.
RoudbariASaghafiF. An evolutionary optimizing approach to neural network architecture for improving identification and modeling of aircraft nonlinear dynamics. Proc IMechE, Part G: J Aerospace Engineering2014; 228: 2178–2191.
13.
AntonakisALoneMCookeA. Neural network based dynamic model and gust identification system for the Jetstream G-NFLA. Proc IMechE, Part G: J Aerospace Engineering2017; 231: 1138–1153.
14.
NiZLiuJWuS, et al.Time-varying state-space model identification of an on-orbit rigid-flexible coupling spacecraft using an improved predictor-based recursive subspace algorithm. Acta Astronaut2019; 163: 157–167.
15.
NiZLiuJWuZ. Identification of the time-varying modal parameters of a spacecraft with flexible appendages using a recursive predictor-based subspace identification algorithm. Proc IMechE, Part G: J Aerospace Engineering2019; 233: 2032–2050.
16.
Yu W, Tan J, Liu CK, et al. Preparing for the unknown: learning a universal policy with online system identification. In: Robotics: science and Systems XIII, Cambridge, MA, 12–16 July 2017.
17.
DongY. An application of deep neural networks to the in-flight parameter identification for detection and characterization of aircraft icing. Aerosp Sci Technol2018; 77: 34–49.
18.
LeCunYBengioYHintonG. Deep learning. Nature2015; 521: 436–444.
19.
Simonyan K and Zisserman A. Very deep convolutional networks for large-scale image recognition. In: Proceedings of the international conference on learning representations, San Diego, CA, 7–9 May 2015.
20.
He K, Zhang X, Ren S, et al. Delving deep into rectifiers: surpassing human-level performance on imagenet classification. In: Proceedings of the IEEE international conference on c–omputer vision, Santiago, Chile, 7–13 December 2015.
21.
JiaFLeiYLinJ, et al.Deep neural networks: a promising tool for fault characteristic mining and intelligent diagnosis of rotating machinery with massive data. Mech Syst Signal Process2016; 72: 303–315.
22.
Cheng L, Wang Z, Song Y, et al. Real-time optimal control for irregular asteroid landings using deep neural networks. arXiv preprint arXiv:1901.02210 2019.
23.
GaoQLiuJJuZ. Robust real-time hand detection and localization for space human–robot interaction based on deep learning. Neurocomputing, Epub ahead of print 24 October 2019. DOI: 10.1016/j.neucom.2019.02.066.
24.
Vincent P, Larochelle H, Bengio Y, et al. Extracting and composing robust features with denoising autoencoders. In: Proceedings of the 25th international conference on machine learning, Helsinki, Finland, 5–9 July 2008.
25.
Glorot X and Bengio Y. Understanding the difficulty of training deep feedforward neural networks. In: Proceedings of the thirteenth international conference on artificial intelligence and statistics, Sardinia, Italy, 13–15 May 2010.
26.
Glorot X, Bordes A and Bengio Y. Deep sparse rectifier neural networks. In: Proceedings of the fourteenth international conference on artificial intelligence and statistics, Fort Lauderdale, FL, 11–13 April 2011.
27.
Sutskever I, Martens J, Dahl G, et al. On the importance of initialization and momentum in deep learning. In: Proceedings of the 30th international conference on machine learning, Atlanta, GA, 16–21 June 2013.
28.
Ioffe S and Szegedy C. Batch normalization: accelerating deep network training by reducing internal covariate shift. In: Proceedings of the 32nd international conference on machine learning, Lille, France, 6–11 July 2015.
29.
BergmannEDzielskiJ. Spacecraft mass property identification with torque-generating control. J Guid Control Dyn1990; 13: 99–103.
30.
Punjani A and Abbeel P. Deep learning helicopter dynamics models. In: 2015 IEEE international conference on robotics and automation, Seattle, WA, 26–30 May 2015.
31.
LeCunYBottouLOrr GB, et al.Efficient backprop. In: GMontavonGBOrrKRMüller (eds). Neural networks: tricks of the trade, 2nd ed. Berlin: Springer, 2012, pp. 9–48.
