Multistep terminal guidance algorithm with intelligent adaptation to wind disturbances
Authors: Klishin A.N., Kolesnikova D.S.
Published in issue: #1(133)/2023
DOI: 10.18698/2308-6033-2023-1-2247
Category: Aviation and Rocket-Space Engineering | Chapter: Aircraft Dynamics, Ballistics, Motion Control
The paper presents a qualitatively new approach to terminal guidance at the final trajectory section for the surface-to-surface class aerial vehicles. The proposed structure of the adaptive control system for an aerial vehicle is based on the multi-step terminal guidance algorithm. Adaptive corrections to the control coefficients were calculated using the developed method for identifying the wind disturbances based on the machine learning models. The work describes technique to form an intelligent algorithm for identifying intensity and direction of the wind load acting on the aerial vehicle in flight. Options of the machine learning models used in the guidance system intelligent block were investigated; their operation results are presented; and the comparative analysis has been carried out. The adaptive guidance system operation procedure is demonstrated on a typical model of the aerial vehicle flying in the atmosphere and targeting a fixed object. Numerical simulation results are presented, and possibility of using such an algorithm and implementing the described system are demonstrated.
References
[1] Klishin A.N. Upravlenie koeffitsientami usileniya stabilizatsii letatelnykh apparatov dlya kompensatsii eye nestatsionarnosti [Control of the gain coefficients of the aircraft stabilization system to compensate for its unsteadiness]. In: Tezisy dokladov XXVII Gagarinskikh chteniy [Abstracts from the XXVII Gagarin Readings]. Moscow, 2002, p. 106.
[2] Klishin A.N., Kolesnikova D.S. High-precision aircraft guidance system with axial acceleration self-tuning. Herald of the Bauman Moscow State Technical University, Series Mechanical Engineering, 2022, no. 4 (143), pp. 60–76. https://doi.org/10.18698/0236-3941-2022-4-60-76
[3] Kuzin S.A. Adaptivnaya sistema upravleniya prodolnym dvizheniem mnogorezhimnogo samoleta [Adaptive longitudinal motion control system of a multi-mode aircraft]. Uchenye zapiski TsAGI — TsAGI Science Journal, 2005, no. 1–2, p. 79.
[4] Zaitsev A.V., Kanushkin S.V., Volkov A.V., Toe Wai Tun. Algoritm optimalnogo upravleniya letatelnogo apparata s uchetom vliyaniya vneshnikh vozmuscheniy [The algorithm of optimal control of the aircraft taking into account the influence of external disturbances]. Transportnoe delo Rossii — Transport Business of Russia, 2015, no. 5, p. 158.
[5] Galaev S.A. Nauchno-metodicheskiy podkhod k otsenivaniyu sostoyaniya slozhnykh obyektov v usloviyakh raznorodnosti izmeritelnoy informatsii [Scientific and methodical approach to state estimation of complex objects under the conditions of heterogeneity of measurement information]. Obrazovatelnye resursy i tekhnologii — Educational Resources and Technologies, 2018, no. 1 (22), pp. 44–48.
[6] Bukov V.N. Adaptivnye prognoziruyuschie sistemy upravleniya poletom [Adaptive predictive flight control systems]. Moscow, Nauka Publ., 1987, 232 p.
[7] Lysenko L.N. Navedenie ballisticheskikh raket [Guidance of ballistic missiles]. Moscow, BMSTU Publ., 2016, 445 p.
[8] Sikharulidze Yu.G. Ballistika i navedenie letatelnykh apparatov [Ballistics and guidance of aircraft]. Moscow, Nauka Publ., 2013, 407 p.
[9] Geron A. Hands-On Machine Learning with Scikit-Learn and TensorFlow: Concepts, Tools, and Techniques for Building Intelligent Systems. O’Reilly Media, Inc., 2019 [In Russ.: Zheron O. Prikladnoe mashinnoe obuchenie s pomoschyu Scikit-Learn i TensorFlow: kontseptsii, instrumenty i tekhniki dlya sozdaniya intellektualnykh sistem. St. Petersburg, Alfa-Kniga Publ., 2018, 688 p.].
[10] Mironov A.M. Mashinnoe obuchenie, chast 1 [Machine learning, part 1]. Moscow, MAKS Press Publ., 2018, 90 p.
[11] Shitikov V.K., Mastitsky S.E. Klassifikatsiya, regressiya, algoritmy Data Mining s ispolzovaniem R [Classification, regression, Data Mining algorithms using R]. Available at: https://github.com/ranalytics/data-mining (accessed December 12, 2021).