Photorealistic simulation platform for autonomous landing of fixed-wing aircraft
Journal of Global Positioning Systems · 2022
DOI: 10.5081/gps.18.2.103 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Global Positioning Systems, 2022. Verfasst von Weili Xue und 3 weiteren Autor:innen.
Abstract (Originalsprache)
To investigate the vision-based autonomous landing of fixed-wing aircraft, we propose a photorealistic simulation platform that leverages ROS, Unreal Engine, and Pixhawk 4. This platform adopts a software-in-the-loop model consisting of rendering, control, communication, and sensor modules. To achieve realistic rendering control, the seawater hydrodynamics and fixed-wing aircraft dynamics are modeled. Based on the platform, we create simulation scenarios under different weather and disturbance conditions for autonomous landing of the aircraft carrier. The platform solves the problem that datasets of related scenes are difficult to collect and experiments are difficult to carry out. To verify the developability of the platform, we design and implement a runway line feature point extraction method (vision-based pose estimation algorithm), and evaluate the performance of the method under various conditions. Experiments show that our software-in-the-loop platform enables vision-based algorithm verification and supports simulation research for the autonomous landing of fixed-wing aircraft.
Zitieren
APA
Xue, W., Sun, Z., Ma, K., & Pei, L. (2022). Photorealistic simulation platform for autonomous landing of fixed-wing aircraft. Journal of Global Positioning Systems. https://doi.org/10.5081/gps.18.2.103
Vancouver
Xue W, Sun Z, Ma K, Pei L. Photorealistic simulation platform for autonomous landing of fixed-wing aircraft. Journal of Global Positioning Systems. 2022. doi:10.5081/gps.18.2.103
BibTeX
@article{Xue2022,
author = {Xue, Weili and Sun, Zhen and Ma, Kehui and Pei, Ling},
title = {Photorealistic simulation platform for autonomous landing of fixed-wing aircraft},
journal = {Journal of Global Positioning Systems},
year = {2022},
doi = {10.5081/gps.18.2.103}
}
EUVIMED