NEURAL NETWORK ALGORITHM SAFE OVERFLIGHT AERIAL OBSTACLES AND PROHIBITED LAND AREAS
https://doi.org/10.26467/2079-0619-2017-20-4-18-24
Abstract
The article presents the algorithm of safe flying around obstacles when making en route flight of manned and unmanned aircraft. The analysis of obstacles in the path of the aircraft is carried out. It is shown that the application of neural networks for this problem solving allows to increase the control system performance and total flight safety. It is proved by modelling. The multilayer network consistent distribution is proposed to be used as neural network structure. In this work a neural network with three layers is used. To solve the problem the aircraft movement in plan is considered. It is important to have data on the Z coordinates of the obstacles vertices. Finally the number of neural network inputs was determined to be four. The number of alternatives, determining the number of neural network outputs is respectively five. As the continuing of the aircraft flight along the original route is possible, as a result, a training sample is in the form of a chart. After training the neural network simulations of its work were made. Obstacles have been formed in advance.
About the Author
D. A. Mikhaylin
Moscow Aviation Institute (national research university)
Russian Federation
Candidate of Technical Sciences, Associate Professor of Moscow Aviation Institute (National Research University)
Russian Federation
Candidate of Technical Sciences, Associate Professor of Moscow Aviation Institute (National Research University)
References
1. Lebedev G.N., Mirzojan L.A. Neyrosetevoe planirovanie dejstviy po obletu nazemnyih obektov gruppoy letatelnyih apparatov [Neural network action planning to fly around ground objects by a group of aircraft]. Aviakosmicheskoe priborostroenie, 2005, no. 12. (in Russian)
2. Lebedev G.N., Rumakina A.V. Neyrosetevoe planirovanie marshruta raznovyisotnogo poleta bespilotnogo letatelnogo apparata [Neural network path planning of a vari-level unmanned aircraft]. Aviakosmicheskoe priborostroenie, 2014, no. 5, pp. 3–8. (in Russian)
3. Lebedev G.N., Rumakina A.V. Sistema logicheskogo upravleniya obhodom prepyatstviy bespilotnyim letatelnyim apparatom pri marshrutnom polete [The logical control system with obstacle avoidance of an unmanned aircraft in flight route]. Elektronnyiy zhurnal «Trudy MAI» [Journal “Trudy MAI”], 2015, no. 83. (in Russian)
4. Kuzin A.V., Kurmakov D.V., Lukyanov A.V., Mihajlin D.A. Neyrosetevaya realizatsiya avtomaticheskogo upravleniya bezopasnoy posadkoy bespilotnogo letatelnogo apparata [The neural network implementation of the automatic control over the safe landing of an unmanned aicraft]. Elektronnyiy zhurnal «Trudy MAI» [Journal “Trudy MAI”], 2013, no. 70. (in Russian)
5. Lebedev G.N., Malygin V.B., Nechaev E.E. Algoritm naznacheniya prioritetnoy ocherednosti sluchayno raspolozhennyim v prostranstve vozdushnyim sudam dlya vyihoda na standartnyiy marshrut pribytiya [The algorithm used to assign the priority sequence randomly located in space aircraft for exit to the standard arrival route]. The Scientific Bulletin of MSTUCA, 2013, no. 198, pp. 31–36. (in Russian)
6. Lebedev G.N., Tin P.Ch., Zo M.T., Medvedev A.V. Upravlenie poletom passazhirskih samoletov pri peresechenii ih marshrutov vo vremya zahoda na posadku [Flight control passenger aircraft crossing their routes during landing]. Jelektronnyj zhurnal «Trudy MAI», 2013, no. 63. (in Russian)
7. Lebedev G.N. et al. Teoriya optimalnyh system [The theory of optimum systems ]. Moscow, MAI, 1993, 317 p. (in Russian)
8. Sobol E.M., Statnikov R.B. Vybor optimalnyh parametrov kompyuternyh system individualnogo obucheniya I povysheniya ih effektivnosti [The selection of the optimum parameters in tasks with many criteria]. Moscow, Science, 1981, 110 p. (in Russian)
9. Lebedev G.N. et al. Planirovanie raboty kompyuternih system individualnogo obucheniya I povysheniya ih effektivnosti [Planning of the work of the computer systems of individual training and an increase in their effectiveness]. The Bulletin of Computer and Information Technologies, 2009, no. 11, 47–54 pp. (in Russian)
10. Zaytsev A.V., Talimanchuk L.L. Intellektualnaya sistema prinyatiya resheniya dlya otsenki nauchnoy deyatelnosti na osnove mnogosigmentnoi sistemy [Intellectual system of decision making for evaluating the scientific activity on the basis of the multisegment system. Neurocomputers, 2008, no. 7, 85–88 pp. (in Russian)
11. Lebedev G.N., Balashov N.M. Intellektualnye sistemy upravleniya [Intellectual systems for control]. M., MGGU, 2000. 117 p. (in Russian)
Review
For citations:
Mikhaylin D.A. NEURAL NETWORK ALGORITHM SAFE OVERFLIGHT AERIAL OBSTACLES AND PROHIBITED LAND AREAS. Civil Aviation High Technologies. 2017;20(4):18-24. (In Russ.) https://doi.org/10.26467/2079-0619-2017-20-4-18-24
Views:
1788
Email this article 
