Aircraft Pilot Assistance System: Detection of Other Aircraft Using Artificial Neural Networks
Artykuł w czasopiśmie
MNiSW
100
Lista 2024
| Status: | |
| Autorzy: | Tomiło Paweł |
| Dyscypliny: | |
| Aby zobaczyć szczegóły należy się zalogować. | |
| Rok wydania: | 2025 |
| Wersja dokumentu: | Elektroniczna |
| Język: | angielski |
| Numer czasopisma: | 2 |
| Wolumen/Tom: | 26 |
| Strony: | 103 - 117 |
| Impact Factor: | 0,9 |
| Web of Science® Times Cited: | 0 |
| Scopus® Cytowania: | 0 |
| Bazy: | Web of Science | Scopus |
| Efekt badań statutowych | NIE |
| Materiał konferencyjny: | NIE |
| Publikacja OA: | TAK |
| Licencja: | |
| Sposób udostępnienia: | Otwarte czasopismo |
| Wersja tekstu: | Ostateczna wersja opublikowana |
| Czas opublikowania: | W momencie opublikowania |
| Data opublikowania w OA: | 17 kwietnia 2025 |
| Abstrakty: | angielski |
| The publication presents a pilot assistance system that is based on video camera frames and uses the developed 3AFPBBYOLO artificial neural network model. The model is able to determine the relative location of other aircraft in the vicinity of an aircraft equipped with this system. Current solutions, the model development process, training, validation, and testing are presented. In the development process, different types of solutions as well as sizes of detection models were tested. In addition, the model inference time with the use of a single-chip mini-computer was examined. The developed model has better values of metrics compared to other solutions. A number of tests were conducted both using a ground station and actual use of the system in flight. The tests conducted using the ground station and the auxiliary hyper-inference algorithm showed that the effective distance from which the model is able to detect an aircraft is 705.09 m. At a fixed distance, the model’s effectiveness is 90.14%. In the flight test using the auxiliary tracking algorithm, the model’s effectiveness was 100%. The developed model is capable of real-time inference. |
