Abstract
Formulated as an optimal control problem, space relative trajectory planning is crucial for on-orbit servicing spacecraft on various missions. While a variety of deep-neural-network (DNN) methods have been proposed to solve the problem, they are energy-consuming and computationally consuming, which limits their on-board deployments. In this work, we proposed a kind of diffractive-deep-neural-network-based optical satellite controller, and transformed the solution of the optimal control problem into a light-field-based fine-grained regression task. Firstly, an electro-optic conversion module was designed to convert numerical relative state variables from electronic signals into light field as the input of a diffractive modulation (DM) module, the diffractive masks of which could be trained to implement complex light field transformation. We used another optic-electro conversion module to convert the light field at the output plane of DM module into electronic signals. Then, we trained the DM module to make the decoded electrical signals consistent with the desired optimal control commands. Therefore, when the light carrying input information and propagating through the well-trained diffractive masks, the DM module could perform diffract-based solution of optimal control problem. The simulation results substantiate the feasibility and effectiveness of our OC-Nets, which can achieve comparable performance to the latest classic DNN methods, except for a few acceptable errors. Different from classic models with much too energy consumption, once fabricated physically, the device of our optical controller can provide optimal control commands at the speed of light, with fairly little computational and energy consumption, and enable the on-board deployment on spacecraft.
This work was supported in part by the National Key Research and Development Program of China under Grant 2020AAA0105500, in part by the National Natural Science Foundation of China under Grant 61971020 and Grant 62031001.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Bennett, T., Schaub, H.: Continuous-time modeling and control using nonsingular linearized relative-orbit elements. J. Guid. Control. Dyn. 39(12), 2605–2614 (2016)
Bogacki, P., Shampine, L.F.: A 3 (2) pair of Runge-Kutta formulas. Appl. Math. Lett. 2(4), 321–325 (1989)
Boyarko, G., Yakimenko, O., Romano, M.: Optimal rendezvous trajectories of a controlled spacecraft and a tumbling object. J. Guid. Control. Dyn. 34(4), 1239–1252 (2011)
Bryson, A.E.: Applied Optimal Control: Optimization, Estimation and Control. CRC Press (1975)
Chang, J., Sitzmann, V., Dun, X., Heidrich, W., Wetzstein, G.: Hybrid optical-electronic convolutional neural networks with optimized diffractive optics for image classification. Sci. Rep. 8(1), 1–10 (2018)
Christodoulides, D.N., Coskun, T.H., Mitchell, M., Segev, M.: Theory of incoherent self-focusing in biased photorefractive media. Phys. Rev. Lett. 78(4), 646 (1997)
Deng, Y., Bao, F., Kong, Y., Ren, Z., Dai, Q.: Deep direct reinforcement learning for financial signal representation and trading. IEEE Trans. Neural Netw. Learn. Syst. 28(3), 653–664 (2017). https://doi.org/10.1109/TNNLS.2016.2522401
Dou, H., Deng, Y., Yan, T., Wu, H., Lin, X., Dai, Q.: Residual D2NN: training diffractive deep neural networks via learnable light shortcuts. Opt. Lett. 45(10), 2688–2691 (2020)
Feng, W., Han, L., Shi, L., Zhao, D., Yang, K.: Optimal control for a cooperative rendezvous between two spacecraft from determined orbits. J. Astronaut. Sci. 63(1), 23–46 (2016)
Gu, Z., Gao, Y., Liu, X.: Optronic convolutional neural networks of multi-layers with different functions executed in optics for image classification. Opt. Express 29(4), 5877–5889 (2021)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016)
Huang, G., Liu, S., Van der Maaten, L., Weinberger, K.Q.: ConDenseNet: an efficient DenseNet using learned group convolutions. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2752–2761 (2018)
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)
Kulce, O., Mengu, D., Rivenson, Y., Ozcan, A.: All-optical information-processing capacity of diffractive surfaces. Light Sci. Appl. 10(1), 1–17 (2021)
LeCun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature 521(7553), 436–444 (2015)
Li, H., Dong, Y., Li, P., Deng, Y.: Optimal real-time approach and capture of uncontrolled spacecraft. J. Spacecr. Rocket. 58(6), 1762–1773 (2021)
Li, W., Gong, S.: Free final-time fuel-optimal powered landing guidance algorithm combing lossless convex optimization with deep neural network predictor. Appl. Sci. 12(7), 3383 (2022)
Li, Y., Chen, R., Sensale-Rodriguez, B., Gao, W., Yu, C.: Real-time multi-task diffractive deep neural networks via hardware-software co-design. Sci. Rep. 11(1), 1–9 (2021)
Lin, X., et al.: All-optical machine learning using diffractive deep neural networks. Science 361(6406), 1004–1008 (2018)
Misra, G., Izadi, M., Sanyal, A., Scheeres, D.: Coupled orbit-attitude dynamics and relative state estimation of spacecraft near small solar system bodies. Adv. Space Res. 57(8), 1747–1761 (2016)
Patterson, M.A., Rao, A.V.: GPOPS-II: a MATLAB software for solving multiple-phase optimal control problems using HP-adaptive Gaussian quadrature collocation methods and sparse nonlinear programming. ACM Trans. Math. Softw. (TOMS) 41(1), 1–37 (2014)
Rahman, M.S.S., Li, J., Mengu, D., Rivenson, Y., Ozcan, A.: Ensemble learning of diffractive optical networks. Light Sci. Appl. 10(1), 1–13 (2021)
Rao, A.V.: A survey of numerical methods for optimal control. Adv. Astronaut. Sci. 135(1), 497–528 (2009)
Scorsoglio, A., D’Ambrosio, A., Ghilardi, L., Gaudet, B., Curti, F., Furfaro, R.: Image-based deep reinforcement meta-learning for autonomous lunar landing. J. Spacecr. Rocket. 59(1), 153–165 (2022)
Shi, J., Chen, Y., Zhang, X.: Broad-spectrum diffractive network via ensemble learning. Opt. Lett. 47(3), 605–608 (2022)
Shi, J., et al: Multiple-view D2NNs array (MDA): realizing robust 3D object recognition (2021)
Starek, J.A., Açıkmeşe, B., Nesnas, I.A., Pavone, M.: Spacecraft autonomy challenges for next-generation space missions. In: Feron, E. (ed.) Advances in Control System Technology for Aerospace Applications. LNCIS, vol. 460, pp. 1–48. Springer, Heidelberg (2016). https://doi.org/10.1007/978-3-662-47694-9_1
Vallado, D.A.: Fundamentals of Astrodynamics and Applications, vol. 12. Springer, Heidelberg (2001)
Veli, M., et al.: Terahertz pulse shaping using diffractive optical networks. In: CLEO: Science and Innovations, pp. SW3K-2. Optical Society of America (2021)
Wang, J., Wu, Y., Liu, M., Yang, M., Liang, H.: A real-time trajectory optimization method for hypersonic vehicles based on a deep neural network. Aerospace 9(4), 188 (2022)
Xing, G.Q., Parvez, S.A.: Alternate forms of relative attitude kinematics and dynamics equations. In: Lynch, J.P. (ed.) 2001 Flight Mechanics Symposium, Greenbelt, MD, USA, pp. 83–97. NASA Goddard Space Flight Center (2001)
Yan, T., et al.: Fourier-space diffractive deep neural network. Phys. Rev. Lett. 123(2), 023901 (2019)
Zhao, Y., Yang, H., Li, S.: Real-time trajectory optimization for collision-free asteroid landing based on deep neural networks. Adv. Space Res. (2022)
Zhou, T., et al.: Large-scale neuromorphic optoelectronic computing with a reconfigurable diffractive processing unit. Nat. Photonics 15(5), 367–373 (2021)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Liu, S., Dou, H., Li, H., Deng, Y. (2022). An Optical Satellite Controller Based on Diffractive Deep Neural Network. In: Fang, L., Povey, D., Zhai, G., Mei, T., Wang, R. (eds) Artificial Intelligence. CICAI 2022. Lecture Notes in Computer Science(), vol 13605. Springer, Cham. https://doi.org/10.1007/978-3-031-20500-2_4
Download citation
DOI: https://doi.org/10.1007/978-3-031-20500-2_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-20499-9
Online ISBN: 978-3-031-20500-2
eBook Packages: Computer ScienceComputer Science (R0)