EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Energy-Aware Multilingual Vision–Language Models for Drone Smart Sensing

J. de Curtò, Mauro Liz, I. de Zarzà, Carlos T. Calafate

Drones · 09.05.2026 · 10 (5) · 361

DOI: 10.3390/drones10050361 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Drones, online erschienen am 09.05.2026. Verfasst von J. de Curtò und 3 weiteren Autor:innen. Die Arbeit nennt 79 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Barcelona Supercomputing Center, Luxembourg Institute of Science and Technology.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Drone-based smart sensing increasingly relies on Vision–Language Models (VLMs) for real-time scene interpretation, obstacle detection, and autonomous navigation reasoning. Deploying such systems at scale demands not only high perceptual accuracy but also energy efficiency, a critical constraint on battery-powered Unmanned Aerial Vehicle (UAV) platforms, and linguistic flexibility for multinational operational contexts. We present a systematic benchmarking framework that jointly evaluates perception performance and inference energy for five open-source VLMs across thirteen languages spanning six language families, including three low-resource varieties (Arabic, Basque, and Luxembourgish). Using imagery sampled from the Berkeley DeepDrive 10K (BDD10K), each model is evaluated on four sensing tasks of increasing difficulty scored via a sentence-transformer backbone, with energy measured following the AI Energy Score methodology (Wh per 1000 queries) through continuous NVML-based GPU power sampling. Across 65 language–model observations, LLaVA-1.6 achieves the highest perception score (S¯=0.160) while Phi-3-Vision attains the best energy efficiency (66.3 Wh/1000 queries); energy consumption and task accuracy are statistically uncorrelated (Spearman ρ=0.001; p=0.995). A formal UAV inference energy model instantiated for four commercial platforms confirms LLaVA-1.6 as Pareto-optimal on heavy-lift platforms (DJI Matrice 300/350 RTK) and LLaVA-1.5 on the energy-constrained Matrice 30; compact UAVs such as the Mavic 3 Enterprise exceed the budget of all evaluated models at standard query rates. Friedman tests reveal significant cross-language variability in energy demands (χ2=40.43; p=3.5×10−8) and navigation reasoning performance (χ2=13.35; p=0.010). Critically, we document a double penalty for low-resource languages, which simultaneously incur higher inference energy costs and lower task accuracy, with direct implications for equitable multilingual UAV deployments.

Förderung

  • Barcelona Supercomputing Center
  • Luxembourg Institute of Science and Technology

Zitieren

APA

Curtò, J. D., Liz, M., Zarzà, I. D., & Calafate, C. T. (2026). Energy-Aware Multilingual Vision–Language Models for Drone Smart Sensing. Drones, 10(5), 361. https://doi.org/10.3390/drones10050361

Vancouver

Curtò JD, Liz M, Zarzà ID, Calafate CT. Energy-Aware Multilingual Vision–Language Models for Drone Smart Sensing. Drones. 2026;10(5):361. doi:10.3390/drones10050361

BibTeX
@article{Curt2026,
  author = {Curtò, J. de and Liz, Mauro and Zarzà, I. de and Calafate, Carlos T.},
  title = {Energy-Aware Multilingual Vision–Language Models for Drone Smart Sensing},
  journal = {Drones},
  year = {2026},
  volume = {10},
  number = {5},
  pages = {361},
  doi = {10.3390/drones10050361}
}
Kostenlos starten →