Nowadays, the use of unmanned aerial vehicles (UAVs) is becoming greater. The growing demand is related to economic considerations, as well as the capability of the UAVs to perform high-risk tasks. Detection of emergencies caused by onboard system failures is one of the main tasks in creating unmanned vehicles. To date, this problem is mainly solved by multiple redundancy. Due to the development of information technologies, intelligent methods of data analysis, namely, artificial neural networks, are becoming increasingly common. Multivariate data analysis using neural networks will allow the level of redundancy to be reduced. It will also allow solving a wide range of tasks in real time. This paper proposes a new approach to developing a comprehensive system for monitoring UAV condition. The approach is based on multivariate data analysis using neural networks. This system is designed to solve a number of tasks such as onboard equipment failure detection based on comprehensive measurement analysis, restoration of readings of faulty sensors, estimation of the aircraft condition, prediction and prevention of hazardous incidents. This system is also capable of detecting control system failure and finish the maneuver by taking over the control. This paper models the solution of these problems using simulated and real flight data.
Using neural networks for monitoring UAV condition
XLIII ACADEMIC SPACE CONFERENCE: dedicated to the memory of academician S.P. Korolev and other outstanding Russian scientists – Pioneers of space exploration ; 2019 ; Moscow, Russia
AIP Conference Proceedings ; 2171 , 1
15.11.2019
6 pages
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
110001 Using neural networks for monitoring UAV condition
TIBKAT | 2019
|Neural networks for engine condition monitoring
British Library Conference Proceedings | 1996
|Artificial Neural Networks in Engine Condition Monitoring
British Library Conference Proceedings | 1997
|SSME Condition Monitoring Using Neural Networks And Plume Spectral Signatures
British Library Conference Proceedings | 1996
|On-line condition monitoring of railway equipment using neural networks
British Library Conference Proceedings | 1995
|