EUVIMEDEUVIMED entdecken
Crossref

AeroVerse: UAV-Agent Benchmark Suite for Simulating, Pre-Training, Finetuning, and Evaluating Aerospace Embodied Foundation Models

Fanglong Yao, Yuanchang Yue, Youzhi Liu, Zhigang Wang, Lei Jin, Bin Zhao, Jian Zhao, Xian Sun, Kun Fu

IEEE Transactions on Pattern Analysis and Machine Intelligence · 2026-10 · 48 (10) · 12301-12318

DOI: 10.1109/tpami.2026.3697634 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2026-10. Verfasst von Fanglong Yao und 8 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 43 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Beijing Natural Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Aerospace embodied intelligence aims to empower uncrewed aerial vehicles (UAVs) and other aerospace platforms to achieve autonomous perception, cognition, and action, as well as egocentric active interaction with humans and the environment. The aerospace embodied foundation model serves as an effective means to realize the autonomous intelligence of UAVs and represents a necessary pathway toward aerospace embodied intelligence. BACKGROUND: However, existing embodied foundation models primarily focus on ground-level intelligent agents in indoor scenarios, while research on UAV intelligent agents remains unexplored, lacking systematic and standardized benchmark suites. AIM: To address this gap, this study aims to construct a comprehensive benchmark suite, AeroVerse, to facilitate the simulation, pre-training, finetuning, and evaluation of aerospace embodied foundation models. INNOVATIONS: We develop AeroSimulator, a simulation platform that encompasses four realistic urban scenes for UAV flight simulation. Additionally, we construct the first large-scale real-world image-text pre-training dataset from a first-person UAV perspective, AerialAgent-Ego15k, and create a virtual image-text-pose alignment dataset, CyberAgent-Ego500k, to facilitate the pre-training of the aerospace embodied foundation model. We clearly define five downstream tasks for the first time, i.e., aerospace embodied scene awareness, spatial reasoning, navigational exploration, task planning, and motion decision, and have constructed corresponding instruction datasets for fine-tuning. We also develop SkyAgent-Eval, a downstream task evaluation system based on GPT-4. Furthermore, we propose SkyAgent, the first UAV-agent large model integrating "perception-reasoning-navigating-planning", which incorporates an aerospace embodied chain-of-thought mechanism and a multitask curriculum learning strategy. RESULTS: By benchmarking ten mainstream models, our results reveal the significant limitations of existing 2D/3D visual-language models in complex aerospace embodied tasks and demonstrate the superior performance of SkyAgent, which outperforms existing methods by an average of 8.52% across four core tasks, underscoring the necessity and contribution of our work.

Förderung

  • National Natural Science Foundation of China
  • Beijing Natural Science Foundation

Zitieren

APA

Yao, F., Yue, Y., Liu, Y., Wang, Z., Jin, L., Zhao, B., Zhao, J., Sun, X., & Fu, K. (2026). AeroVerse: UAV-Agent Benchmark Suite for Simulating, Pre-Training, Finetuning, and Evaluating Aerospace Embodied Foundation Models. IEEE Transactions on Pattern Analysis and Machine Intelligence, 48(10), 12301-12318. https://doi.org/10.1109/tpami.2026.3697634

Vancouver

Yao F, Yue Y, Liu Y, Wang Z, Jin L, Zhao B, et al. AeroVerse: UAV-Agent Benchmark Suite for Simulating, Pre-Training, Finetuning, and Evaluating Aerospace Embodied Foundation Models. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2026;48(10):12301-12318. doi:10.1109/tpami.2026.3697634

BibTeX
@article{Yao2026,
  author = {Yao, Fanglong and Yue, Yuanchang and Liu, Youzhi and Wang, Zhigang and Jin, Lei and Zhao, Bin and Zhao, Jian and Sun, Xian and Fu, Kun},
  title = {AeroVerse: UAV-Agent Benchmark Suite for Simulating, Pre-Training, Finetuning, and Evaluating Aerospace Embodied Foundation Models},
  journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year = {2026},
  volume = {48},
  number = {10},
  pages = {12301--12318},
  doi = {10.1109/tpami.2026.3697634}
}
Kostenlos starten →