Deep Learning Architecture for UAV Traffic-Density Prediction

• Published in: Drones (Basel) Vol. 7; no. 2; p. 78
• Main Authors: Alharbi, Abdulrahman, Petrunin, Ivan, Panagiotakopoulos, Dimitrios
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.01.2023
• Subjects: Air traffic control; Air traffic management; Aircraft; Airspace; Algorithms; Alliances; Altitude; Artificial neural networks; Civil aviation; Coders; Complexity; complexity metrics; Deep learning; Delivery services; Density; Drone aircraft; Drones; Efficiency; Emergency procedures; Encoders-Decoders; Inspection; long short-term memory (LSTM) networks; Machine learning; Neural networks; Post offices; Prediction models; Radio communications; Regression models; Systems stability; Traffic flow; Unmanned aerial vehicles; unmanned aerial vehicles (UAVs); Unmanned aircraft; unmanned traffic management (UTM); Vehicles; Weather; United Kingdom; United Kingdom > UK
• ISSN: 2504-446X; 2504-446X
• DOI: 10.3390/drones7020078

The research community has paid great attention to the prediction of air traffic flows. Nonetheless, research examining the prediction of air traffic patterns for unmanned aircraft traffic management (UTM) is relatively sparse at present. Thus, this paper proposes a one-dimensional convolutional neural network and encoder-decoder LSTM framework to integrate air traffic flow prediction with the intrinsic complexity metric. This adapted complexity metric takes into account the important differences between ATM and UTM operations, such as dynamic flow structures and airspace density. Additionally, the proposed methodology has been evaluated and verified in a simulation scenario environment, in which a drone delivery system that is considered essential in the delivery of COVID-19 sample tests, package delivery services from multiple post offices, an inspection of the railway infrastructure and fire-surveillance tasks. Moreover, the prediction model also considers the impacts of other significant factors, including emergency UTM operations, static no-fly zones (NFZs), and variations in weather conditions. The results show that the proposed model achieves the smallest RMSE value in all scenarios compared to other approaches. Specifically, the prediction error of the proposed model is 8.34% lower than the shallow neural network (on average) and 19.87% lower than the regression model on average.