Abstract This work is devoted to generating optimal guidance commands in real time for attitude-constrained solar sailcrafts in coplanar circular-to-circular interplanetary transfers. Firstly, a nonlinear optimal control problem is established, and the necessary conditions for optimality are derived by Pontryagin’s Minimum Principle. Under some mild assumptions, the attitude constraints are rewritten as control constraints, which are replaced by a saturation function so that a parameterized system is formulated. This allows one to generate an optimal trajectory via solving an initial value problem, making it efficient to collect a dataset containing optimal samples, which are essential for training Neural Networks (NNs) to achieve real-time implementation. However, the optimal guidance command may suddenly change from one extreme to another, resulting in discontinuous jumps that generally impair the NN’s approximation performance. To address this issue, we use two co-states that the optimal guidance command depends on, to detect discontinuous jumps. A procedure for preprocessing these jumps is then established, thereby ensuring that the preprocessed guidance command remains smooth everywhere. Meanwhile, the sign of one co-state is found to be sufficient to revert the preprocessed guidance command back into the original optimal guidance command. Furthermore, three NNs are built and trained offline, and they cooperate to precisely generate the optimal guidance command in real time. Finally, numerical simulations are presented to demonstrate the developments of the paper.

    Highlights New dataset generation method for training neural networks used in optimal guidance. A preprocessing method ensuring the smoothness of the preprocessed guidance command. Using the sign of one costate to convert the preprocessed guidance command. Improved fitting performance enabling the precise generation of the guidance command.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Real-time optimal control for attitude-constrained solar sailcrafts via neural networks


    Beteiligte:
    Wang, Kun (Autor:in) / Lu, Fangmin (Autor:in) / Chen, Zheng (Autor:in) / Li, Jun (Autor:in)

    Erschienen in:

    Acta Astronautica ; 216 ; 446-458


    Erscheinungsdatum :

    2024-01-18


    Format / Umfang :

    13 pages




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Initialization of linear controllers for constrained time-optimal attitude control

    Vigneron, Adam / Delchambre, Simon / Ziegler, Tobias et al. | AIAA | 2022


    Initialization of Linear Controllers for Constrained Time-Optimal Attitude Control

    Vigneron, Adam / Delchambre, Simon / Ziegler, Tobias et al. | TIBKAT | 2022


    Artificial neural networks in Space Station optimal attitude control

    Kumar, R. R. / Seywald, H. / Deshpande, S. M. et al. | British Library Conference Proceedings | 1995


    SOC-i: A CubeSat Demonstration of Optimization-Based Real-Time Constrained Attitude Control

    Reynolds, Taylor P. / Kelly, Charles L. / Morgan, Cole et al. | IEEE | 2021


    Real-Time Computations of Time-Optimal Magnetic Attitude Control (AAS 05-233)

    Yan, H. / Fleming, A. / Ross, I. M. et al. | British Library Conference Proceedings | 2005