Synthesis of an intelligent algorithm for assessing the orientation of movable objects of transport infrastructure on the basis of a multiparameter neural network identifier
https://doi.org/10.46973/0201-727X_2023_1_144
Abstract
General measurement informational processing algorithms do not always provide the necessary accuracy of assessment under external influences and require large computing power, which is difficult to implement under conditions of restrictions on the weight and size characteristics of UAVs transport. The use of algorithms for dynamic estimation of the UAV angular orientation based on an adaptive model in combination with the use of multilayer neural networks of direct propagation makes it possible to reduce the error in estimating the parameters of a dynamic system without a significant increase in computational costs. The paper presents the synthesis of a UAV orientation estimation system with a multiparameter neural network identifier, which makes it possible to improve the estimation accuracy in comparison with the classical Kalman filter.
About the Authors
A. A. Kostoglotov
Rostov State Transport University (RSTU)
Russian Federation
Russian Federation
Kostoglotov Andrey Alexandrovich, Research Department, Chair «Communication on Railway Transport», Doctor of Engineering Sciences, Professor, Leading Researcher
S. V. Lazarenko
Rostov State Transport University (RSTU)
Russian Federation
Russian Federation
Lazarenko Sergey Valeryevich, Research Department, Chair «Communication on Railway Transport», Candidate of Engineering Sciences, Associate Professor
A. S. Penkov
Rostov State Transport University (RSTU)
Russian Federation
Russian Federation
Penkov Anton Sergeevich, Research Department, Chair «Communication on Railway Transport», Researcher
V. O. Zekhtser
Rostov State Transport University (RSTU)
Russian Federation
Russian Federation
Zekhtser Vladimir Olegovich, Research Department, Chair «Communication on Railway Transport», Junior Researcher
Kh. Sh. Kulbikayan
Rostov State Transport University (RSTU)
Russian Federation
Russian Federation
Kulbikayan Khacheres Shagenovich, Chair «Communication on Railway Transport», Candidate of Engineering Sciences, Associate Professor
References
1. Grigorenko, N. I. Diagnostics of the road condition using an unmanned aerial vehicle / N. I. Grigorenko, E. E. Yanchuk // World of Transport. – 2017. – Vol.. 15, No. 3 (70). – P. 86–92. – EDN ZTPSBB.
2. Pavlov, V. M. Measuring the angular position of a vehicle / V. M. Pavlov // Transport : science, education, production. – Rostov-onDon : RSTU, 2022. – Vol. 1. – P. 209–212. – EDN JLMBMP.
3. Kostoglotov, A. A. Synthesis of adaptive algorithms for assessing the orientation of unmanned vehicles using an intelligent neural network identifier / A. A. Kostoglotov, A. S. Penkov, V. O. Zekhtser // Vestnik Rostovskogo Gosudarstvennogo Universiteta Putey Soobshcheniya – 2022. – No. 1 (85). – P. 186–194. – DOI 10.46973/0201-727X_2022_1_186.
4. Analytical review of modern intelligent information technologies in engineering and in production / S. M. Kovalev [et al.] // Vestnik Rostovskogo Gosudarstvennogo Universiteta Putey Soobshcheniya. – 2019. – No. 1 (73). – P. 60–75. – EDN ZBKLIL.
5. Chulin, N. A. Unmanned aerial vehicle control system / N. A. Chulin, I. V. Mironova // Engineering Journal: Science and Innovation. – 2018. – No. 9. – P. 1–11. – EDN VAGNVQ.
6. Gutsevich, D. E. Modeling the behavior of an aircraft-type aircraft with automatic control in various flight modes / D. E. Gutsevich // Mathematical modeling, computer and field experiment in natural sciences. – 2018. – No. 1. – P. 12–23. – EDN XZCQWT.
7. Konovalov, A. A. Fundamentals of trajectory processing of radar information. Part 2 / A. A. Konovalov. – St. Petersburg : SPbSETU «LETI», 2013. – 180 p. – ISBN 978-5-7629-1544-1.
8. Kostoglotov, A. A. Structural-parametric synthesis of a tracking filter based on decomposition by target functional with adaptation to trajectory perturbations / A. A. Kostoglotov, A. S. Penkov, S. V. Lazarenko // Information-measuring and control systems. – 2021. – Vol. 19. – No. 2. – P. 14–25. – DOI 10.18127/j20700814-202102-02. – EDN FWFWBO.
9. Kostoglotov, A. A. Method of synthesis of adaptive algorithms for estimating parameters of dynamic systems based on the principle of decomposition and methodology of the combined maximum principle / A. A. Kostoglotov, A. S. Penkov, S. V. Lazarenko // News of higher educational institutions. The North Caucasus region. Series : Natural Sciences. – 2020. – No. 4 (208). – P. 22–28. – EDN QOAKMD.
10. Sokolov, S. Adaptive stochastic filtration based on the estimation of the covariance matrix of measurement noises using irregular accurate observations / S. Sokolov, M. Polyakova, A. Novikov // Inventions. – 2021. – Vol. 6. No. 1. – P. 1–12. – EDN KOPLQN.
11. Matveev, V. V. Fundamentals of construction of free-form inertial navigation systems / V. V. Matveev, V. Ya. Raspopov // JSC «Concern "Central Research Institute "Electropribor"». – 2009. – 280 p. – EDN QNWJZT.
12. Kovalev, V. S. Adaptive network models of multisensory data fusion in hybrid diagnostic systems / V. S. Kovalev, S. M. Kovalev, A. V. Sukhanov // Vestnik Rostovskogo Gosudarstvennogo Universiteta Putey Soobshcheniya. – 2020. – No. 1 (77). – P. 153– 162. – EDN ROSZKE.
13. Pupkov, K. A. Methods of robust, neurofuzzy and adaptive control : A cycle of textbooks and manuals "Methods of the theory of automatic control" / K. A. Pupkov, N. D. Egupov, A. I. Gavrilov // Moscow : Bauman Moscow State Technical University (National Research University). – 2001. – 744 p. – EDN WEYIDX.
14. Egorchev, M. V. Neural network semi– empirical approach to modeling longitudinal motion and identification of aerodynamic characteristics of a maneuverable aircraft / M. V. Egorchev, Yu. V. Tyumintsev // Proceedings of MAI. – 2017. – 31 p. – EDN YZMBEN.
Review
For citations:
Kostoglotov A.A., Lazarenko S.V., Penkov A.S., Zekhtser V.O., Kulbikayan Kh.Sh. Synthesis of an intelligent algorithm for assessing the orientation of movable objects of transport infrastructure on the basis of a multiparameter neural network identifier. Vestnik Rostovskogo Gosudarstvennogo Universiteta Putej Soobshcheniya. 2023;(1):144-151. (In Russ.) https://doi.org/10.46973/0201-727X_2023_1_144
Views:
6
JATS XML
Email this article 