Deep Learning Architecture for UAV Traffic-Density Prediction
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 neu...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-01-01
|
| Series: | Drones |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2504-446X/7/2/78 |
| _version_ | 1797621453889208320 |
|---|---|
| author | Abdulrahman Alharbi Ivan Petrunin Dimitrios Panagiotakopoulos |
| author_facet | Abdulrahman Alharbi Ivan Petrunin Dimitrios Panagiotakopoulos |
| author_sort | Abdulrahman Alharbi |
| collection | DOAJ |
| description | 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. |
| first_indexed | 2024-03-11T08:56:09Z |
| format | Article |
| id | doaj.art-a8668e4eed6a46eca9ef9feb79045783 |
| institution | Directory Open Access Journal |
| issn | 2504-446X |
| language | English |
| last_indexed | 2024-03-11T08:56:09Z |
| publishDate | 2023-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Drones |
| spelling | doaj.art-a8668e4eed6a46eca9ef9feb790457832023-11-16T20:06:15ZengMDPI AGDrones2504-446X2023-01-01727810.3390/drones7020078Deep Learning Architecture for UAV Traffic-Density PredictionAbdulrahman Alharbi0Ivan Petrunin1Dimitrios Panagiotakopoulos2School of Aerospace, Transport and Manufacturing, Cranfield University, Bedford MK43 0AL, UKSchool of Aerospace, Transport and Manufacturing, Cranfield University, Bedford MK43 0AL, UKSchool of Aerospace, Transport and Manufacturing, Cranfield University, Bedford MK43 0AL, UKThe 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.https://www.mdpi.com/2504-446X/7/2/78complexity metricslong short-term memory (LSTM) networksunmanned aerial vehicles (UAVs)unmanned traffic management (UTM) |
| spellingShingle | Abdulrahman Alharbi Ivan Petrunin Dimitrios Panagiotakopoulos Deep Learning Architecture for UAV Traffic-Density Prediction Drones complexity metrics long short-term memory (LSTM) networks unmanned aerial vehicles (UAVs) unmanned traffic management (UTM) |
| title | Deep Learning Architecture for UAV Traffic-Density Prediction |
| title_full | Deep Learning Architecture for UAV Traffic-Density Prediction |
| title_fullStr | Deep Learning Architecture for UAV Traffic-Density Prediction |
| title_full_unstemmed | Deep Learning Architecture for UAV Traffic-Density Prediction |
| title_short | Deep Learning Architecture for UAV Traffic-Density Prediction |
| title_sort | deep learning architecture for uav traffic density prediction |
| topic | complexity metrics long short-term memory (LSTM) networks unmanned aerial vehicles (UAVs) unmanned traffic management (UTM) |
| url | https://www.mdpi.com/2504-446X/7/2/78 |
| work_keys_str_mv | AT abdulrahmanalharbi deeplearningarchitectureforuavtrafficdensityprediction AT ivanpetrunin deeplearningarchitectureforuavtrafficdensityprediction AT dimitriospanagiotakopoulos deeplearningarchitectureforuavtrafficdensityprediction |