aiqbench analysis
This paper introduces AeroIntent, a simulation-based benchmark for evaluating mission-level intent recognition in unmanned aerial vehicles using synchronized multimodal sensor data. The authors propose AeroMIR, a structured reasoning model that significantly outperforms existing baselines by integrating radar trajectories with visual evidence to handle incomplete histories and unseen tactical scenarios. This work advances the field of autonomous systems by providing a rigorous framework for assessing AI capabilities in complex surveillance and security contexts.
AeroIntent: A Synchronized Multimodal Benchmark for UAV Mission Intent Recognition
arXiv:2610.25869v1 Announce Type: new Abstract: Reliable recognition of mission-level UAV intent from multimodal observations is essential for low-altitude airspace surveillance and security. Existing multimodal UAV datasets primarily support detection, tracking, and localization, leaving limited support for evaluating mission-intent recognition across held-out trajectory programs and incomplete radar histories. We introduce AeroIntent, a simulation-based benchmark comprising 990 synchronized multimodal sequences generated by 22 trajectory programs across five mission intents. The benchmark provides radar, RGB, infrared, and depth observations, together with program-disjoint, radar-gap, and observation-prefix evaluation protocols. Building on this benchmark, we propose AeroMIR, a structured Multimodal Intent Reasoner that integrates radar trajectories, derived behavioral features, selected visual evidence, and task knowledge for candidate-based UAV intent recognition. Under a strict program-disjoint protocol evaluating unseen tactical realizations, AeroMIR achieves a mean accuracy of 71.56% and a mean Macro-F1 of 69.84% across three seeds, outperforming the strongest trajectory-only baseline by 7.79 Macro-F1 points and a conventional radar-RGB late-fusion baseline by 19.32 points. AeroMIR also demonstrates strong performance under controlled radar gaps and early intent recognition settings. AeroIntent and AeroMIR provide a reproducible foundation for multimodal UAV intent recognition. The benchmark is available at: https://anonymous.4open.science/r/AeroIntent
Abstract:Reliable recognition of mission-level UAV intent from multimodal observations is essential for low-altitude airspace surveillance and security. Existing multimodal UAV datasets primarily support detection, tracking, and localization, leaving limited support for evaluating mission-intent recognition across held-out trajectory programs and incomplete radar histories. We introduce AeroIntent, a simulation-based benchmark comprising 990 synchronized multimodal sequences generated by 22 trajectory programs across five mission intents. The benchmark provides radar, RGB, infrared, and depth observations, together with program-disjoint, radar-gap, and observation-prefix evaluation protocols. Building on this benchmark, we propose AeroMIR, a structured Multimodal Intent Reasoner that integrates radar trajectories, derived behavioral features, selected visual evidence, and task knowledge for candidate-based UAV intent recognition. Under a strict program-disjoint protocol evaluating unseen tactical realizations, AeroMIR achieves a mean accuracy of 71.56% and a mean Macro-F1 of 69.84% across three seeds, outperforming the strongest trajectory-only baseline by 7.79 Macro-F1 points and a conventional radar-RGB late-fusion baseline by 19.32 points. AeroMIR also demonstrates strong performance under controlled radar gaps and early intent recognition settings. AeroIntent and AeroMIR provide a reproducible foundation for multimodal UAV intent recognition. The benchmark is available at: this https URL
Robotics (cs.RO)
arXiv:2610.25869 [cs.RO]
(or arXiv:2610.25869v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2610.25869 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 279 words
Related coverage
GhostPrefix: Synchronization-Free Universal Prefix Attacks for ASR Manipulation
arXiv · Nov 2, 2026
Ex-Omni: Enabling 3D Facial Animation Generation for Omni-modal Large Language Models
arXiv · Sep 5, 2026
Empathy as Predictive Misalignment Tolerance: A Co-Regulation Framework and the Regime Structure of Dialogue Repair
arXiv · Jul 20, 2026
BioKD: Selective Physiology-to-Video Knowledge Distillation via Reliability Gate for Emotion Recognition
arXiv · Aug 7, 2026