EUVIMEDEUVIMED entdecken
Crossref

Research on Autonomous Obstacle Avoidance for Quadrotor UAVs Based on Deep Learning

Qiyan Song

Journal of Intelligence and Knowledge Engineering · 2026-06 · 4 (2) · 149-156

DOI: 10.62517/jike.202604221 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Intelligence and Knowledge Engineering, 2026-06. Verfasst von Qiyan Song. Die Arbeit nennt 13 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

To address single‑sensor limitations and real‑time constraints in quadrotor UAV obstacle avoidance, this paper proposes a closed‑loop framework integrating cross‑modal RGB‑D/radar fusion, PPO‑based dual‑stream decision‑making with online incremental learning, and differential‑flatness trajectory generation. Under ideal conditions, detection accuracy reaches 95.4%, avoidance success rate 91.8%, and end‑to‑end latency 109 ms; under dark and fully occluded conditions, accuracy is 89.2% and avoidance success rate drops to 82.0%, indicating a performance boundary. These results validate the framework's potential in simulated unstructured dynamic environments while revealing areas for improvement under extreme perception degradation.

Zitieren

APA

Song, Q. (2026). Research on Autonomous Obstacle Avoidance for Quadrotor UAVs Based on Deep Learning. Journal of Intelligence and Knowledge Engineering, 4(2), 149-156. https://doi.org/10.62517/jike.202604221

Vancouver

Song Q. Research on Autonomous Obstacle Avoidance for Quadrotor UAVs Based on Deep Learning. Journal of Intelligence and Knowledge Engineering. 2026;4(2):149-156. doi:10.62517/jike.202604221

BibTeX
@article{Song2026,
  author = {Song, Qiyan},
  title = {Research on Autonomous Obstacle Avoidance for Quadrotor UAVs Based on Deep Learning},
  journal = {Journal of Intelligence and Knowledge Engineering},
  year = {2026},
  volume = {4},
  number = {2},
  pages = {149--156},
  doi = {10.62517/jike.202604221}
}
Kostenlos starten →