Article · International Review of Aerospace Engineering · June 30, 2024
The Use of Artificial Intelligence to Increase the Functional Stability of UAV Systems
Aigerim T. Mazakova, Sholpan A. Jomartova, Talgat Zh. Mazakov, Rashit Brzhanov, Dmitry A. Gura
Abstract
Automated troubleshooting of Unmanned Aerial Vehicles (UAVs) is essential in modern aviation. With the growing number of UAVs across various fields, fault detection and prediction have become essential for ensuring flight safety, enhancing reliability, and reducing operating costs. This article aims to evaluate various machine learning methods, including neural networks. The article analyzes the effectiveness of several machine-learning methods in detecting various types of malfunctions in UAVs, including engine failures and issues with rudders, elevators, and ailerons. Methods such as CNN-LSTM, CNN-NB, and CNN-LSTM-NB were tested on relevant data on emulated UAV flights with malfunctions. The results of the study revealed that different machine-learning methods vary in their effectiveness in detecting malfunctions in UAVs. For example, the CNN-LSTM method proved to be the most effective for detecting engine failures, while the CNN-LSTM-NB method showed high accuracy in detecting rudder and aileron failures. These findings can be useful for improving maintenance and safety systems in UAV aviation.
More by Sholpan Jomartova
- MODIFIED IMAGE ENCRYPTION ALGORITHM BASED ON CHAOTIC DYNAMICS2026
- Optimization of the production process of a bakery enterprise based on the CP-SAT method2026
- Mathematical model of population evacuation taking into account time characteristicsBulletin of the National Engineering Academy of the Republic of Kazakhstan · 2026
- Validation-Generalization Asymmetry in Metaheuristic CNN Forecasting: An Extremely Small Time-Series CaseSymmetry · 2026 · Citations: 2