Closed-loop deep neural network optimal control algorithm and error analysis for powered landing under uncertainties

Wenbo Li; Yu Song; Lin Cheng; Shengping Gong

doi:10.1007/s42064-022-0153-1

Astrodynamics 2023, 7(2): 211-228 https://doi.org/10.1007/s42064-022-0153-1

Research Article | Issue | Published: 23 November 2022

Closed-loop deep neural network optimal control algorithm and error analysis for powered landing under uncertainties

Show Author's Information Hide Author's Information Wenbo Li^¹, Yu Song^¹, Lin Cheng^², Shengping Gong^²(

)

School of Aerospace Engineering, Tsinghua University, Beijing 100084, China

School of Astronautics, Beihang University, Beijing 100191, China

Keywords:

deep neural network (DNN), model predictive control (MPC), powered landing guidance, linear covariance analysis (LCA)

Cite this article:

Li W, Song Y, Cheng L, et al. Closed-loop deep neural network optimal control algorithm and error analysis for powered landing under uncertainties. Astrodynamics, 2023, 7(2): 211-228. https://doi.org/10.1007/s42064-022-0153-1

Download citation

EndNote(RIS)

BibTeX

6014

Views

Downloads

Citations

Crossref

WoS

Scopus

CSCD

Abstract Full text About this article

Abstract

Real-time guidance is critical for the vertical recovery of rockets. However, traditional sequential convex optimization algorithms suffer from shortcomings in terms of their poor real-time performance. This work focuses on applying the deep learning-based closed-loop guidance algorithm and error propagation analysis for powered landing, thereby significantly improving the real-time performance. First, a controller consisting of two deep neural networks is constructed to map the thrust direction and magnitude of the rocket according to the state variables. Thereafter, the analytical transition relationships between different uncertainty sources and the state propagation error in a single guidance period are analyzed by adopting linear covariance analysis. Finally, the accuracy of the proposed methods is verified via a comparison with the indirect method and Monte Carlo simulations. Compared with the traditional sequential convex optimization algorithm, our method reduces the computation time from 75 ms to less than 1 ms. Therefore, it shows potential for online applications.

Full text

Abstract

Full text

Outline

About this article

Closed-loop deep neural network optimal control algorithm and error analysis for powered landing under uncertainties

Show Author's information Hide Author's Information Wenbo Li^¹, Yu Song^¹, Lin Cheng^², Shengping Gong^²(

)

School of Aerospace Engineering, Tsinghua University, Beijing 100084, China

School of Astronautics, Beihang University, Beijing 100191, China

Abstract

Keywords: deep neural network (DNN), model predictive control (MPC), powered landing guidance, linear covariance analysis (LCA)

References(36)

[1]

Song, Z. Y., Wang, C., Theil, S., Seelbinder, D., Sagliano, M., Liu, X. F., Shao, Z. J. Survey of autonomous guidance methods for powered planetary landing. Frontiers of Information Technology & Electronic Engineering, 2020, 21(5): 652–674.

DOI Google Scholar

[2]

Baiocco, P. Overview of reusable space systems with a look to technology aspects. Acta Astronautica, 2021, 189: 10–25.

DOI Google Scholar

[3]

Connor, M. A. Gravity turn trajectories through the atmosphere. Journal of Spacecraft and Rockets, 1966, 3(8): 1308–1311.

DOI Google Scholar

[4]

Culler, G. J., Fried, B. D. Universal gravity turn trajectories. Journal of Applied Physics, 1957, 28(6): 672–676.

DOI Google Scholar

[5]

Klumpp, A. R. Apollo lunar descent guidance. Automatica, 1974, 10(2): 133–146.

DOI Google Scholar

[6]

Ingoldby, R. N. Guidance and control system design of the Viking planetary lander. Journal of Guidance and Control, 1978, 1(3): 189–196.

DOI Google Scholar

[7]

Sostaric, R., Rea, J. Powered descent guidance methods for the moon and Mars. In: Proceedings of the AIAA Guidance, Navigation, and Control Conference and Exhibit, 2005: AIAA 2005-6287.

DOI

[8]

Cheng, R. K., Conrad, D. A., Meredith, C. M. Design considerations for Surveyor guidance. Journal of Spacecraft and Rockets, 1966, 3(11): 1569–1576.

DOI Google Scholar

[9]

Wong, E. C., Singh, G., Masciarelli, J. P. Guidance and control design for hazard avoidance and safe landing on Mars. Journal of Spacecraft and Rockets, 2006, 43(2): 378–384.

DOI Google Scholar

[10]

Topcu, U., Casoliva, J., Mease, K. Fuel efficient powered descent guidance for Mars landing. In: Proceedings of the AIAA Guidance, Navigation, and Control Conference and Exhibit, 2005: AIAA 2005-6286.

DOI

[11]

Rea, J., Bishop, R. Analytical dimensional reduction of a fuel optimal powered descent subproblem. In: Proceedings of the AIAA Guidance, Navigation, and Control Conference, 2010: AIAA 2010-8026.

DOI

[12]

Najson, F., Mease, K. A computationally non-expensive guidance algorithm for fuel efficient soft landing. In: Proceedings of the AIAA Guidance, Navigation, and Control Conference and Exhibit, 2005: AIAA 2005-6289.

DOI

[13]

Lu, P. Augmented Apollo powered descent guidance. Journal of Guidance, Control, and Dynamics, 2018, 42(3): 447–457.

DOI Google Scholar

[14]

Lu, P. Propellant-optimal powered descent guidance. Journal of Guidance, Control, and Dynamics, 2017, 41(4): 813–826.

DOI Google Scholar

[15]

Acikmese, B., Ploen, S. R. Convex programming approach to powered descent guidance for Mars landing. Journal of Guidance, Control, and Dynamics, 2007, 30(5): 1353–1366.

DOI Google Scholar

[16]

Blackmore, L., Açikmeşe, B., Scharf, D. P. Minimum-landing-error powered-descent guidance for Mars landing using convex optimization. Journal of Guidance, Control, and Dynamics, 2010, 33(4): 1161–1171.

DOI Google Scholar

[17]

Liu, X. F. Autonomous trajectory planning by convex optimization. Ph. D. Dissertation. Iowa State University, 2013.

[18]

Liu, X. F., Lu, P., Pan, B. F. Survey of convex optimization for aerospace applications. Astrodynamics, 2017, 1(1): 23–40.

DOI Google Scholar

[19]

Song, Y., Miao, X. Y., Gong, S. P. Adaptive powered descent guidance based on multi-phase pseudospectral convex optimization. Acta Astronautica, 2021, 180: 386–397.

DOI Google Scholar

[20]

Song, Z. Y., Wang, C. Powered soft landing guidance method for launchers with non-cluster configured engines. Acta Astronautica, 2021, 189: 379–390.

DOI Google Scholar

[21]

Izzo, D., Märtens, M., Pan, B. F. A survey on artificial intelligence trends in spacecraft guidance dynamics and control. Astrodynamics, 2019, 3(4): 287–299.

DOI Google Scholar

[22]

Chai, R. Q., Tsourdos, A., Savvaris, A., Chai, S. C., Xia, Y. Q., Philip Chen, C. L. Review of advanced guidance and control algorithms for space/aerospace vehicles. Progress in Aerospace Sciences, 2021, 122: 100696.

DOI Google Scholar

[23]

Shirobokov, M., Trofimov, S., Ovchinnikov, M. Survey of machine learning techniques in spacecraft control design. Acta Astronautica, 2021, 186: 87–97.

DOI Google Scholar

[24]

Sánchez-Sánchez, C., Izzo, D. Real-time optimal control via deep neural networks: Study on landing problems. Journal of Guidance, Control, and Dynamics, 2018, 41(5): 1122–1135.

DOI Google Scholar

[25]

Cheng, L., Wang, Z. B., Song, Y., Jiang, F. H. Real-time optimal control for irregular asteroid landings using deep neural networks. Acta Astronautica, 2020, 170: 66–79.

DOI Google Scholar

[26]

Cheng, L., Wang, Z. B., Jiang, F. H. Real-time control for fuel-optimal Moon landing based on an interactive deep reinforcement learning algorithm. Astrodynamics, 2019, 3(4): 375–386.

DOI Google Scholar

[27]

Cheng, L., Li, H. N., Wang, Z. W., Jiang, F. H. Fast solution continuation of time-optimal asteroid landing trajectories using deep neural networks. Acta Astronautica, 2020, 167: 63–72.

DOI Google Scholar

[28]

Song, Y., Miao, X. Y., Cheng, L., Gong, S. P. The feasibility criterion of fuel-optimal planetary landing using neural networks. Aerospace Science and Technology, 2021, 116: 106860.

DOI Google Scholar

[29]

Gaudet, B., Furfaro, R. Adaptive pinpoint and fuel efficient Mars landing using reinforcement learning. IEEE/CAA Journal of Automatica Sinica, 2014, 1(4): 397–411.

DOI Google Scholar

[30]

Gaudet, B., Linares, R., Furfaro, R. Deep reinforcement learning for six degree-of-freedom planetary landing. Advances in Space Research, 2020, 65(7): 1723–1741.

DOI Google Scholar

[31]

Geller, D. K. Linear covariance techniques for orbital rendezvous analysis and autonomous onboard mission planning. Journal of Guidance, Control, and Dynamics, 2006, 29(6): 1404–1414.

DOI Google Scholar

[32]

Geller, D. K., Rose, M. B., Woffinden, D. C. Event triggers in linear covariance analysis with applications to orbital rendezvous. Journal of Guidance, Control, and Dynamics, 2009, 32(1): 102–111.

DOI Google Scholar

[33]

Geller, D. K., Christensen, D. P. Linear covariance analysis for powered lunar descent and landing. Journal of Spacecraft and Rockets, 2009, 46(6): 1231–1248.

DOI Google Scholar

[34]

Jang, J., Bhatt, S., Fritz, M., Woffinden, D. Linear covariance analysis for a lunar lander. In: Proceedings of the AIAA Guidance, Navigation, and Control Conference, 2017: AIAA 2017-1499.

DOI

[35]

Christensen, R., Geller, D., Hansen, M. Linear covariance navigation analysis of range and image measurement processing for autonomous lunar lander missions. In: Proceedings of the IEEE/ION Position, Location and Navigation Symposium, 2020: 1524–1535.

DOI

[36]

Li, W. B., Gong, S. P. Free final-time fuel-optimal powered landing guidance algorithm combing lossless convex optimization with deep neural network predictor. Applied Sciences, 2022, 12(7): 3383.

DOI Google Scholar

About this article

Publication history

Acknowledgements

Publication history

Received: 11 July 2022

Accepted: 04 September 2022

Published: 23 November 2022

Issue date: June 2023

Copyright

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant Nos. 11822205 and 11772167).