U2UData-2: A Scalable Swarm UAVs Autonomous Flight Dataset for Long-horizon Tasks

2025-08-25

AI Summary

Existing approaches are hindered by data scarcity, limiting the practical deployment of swarm UAVs in long-horizon (LH) missions—which demand modeling of long-term dependencies, persistent state maintenance, and dynamic target adaptation. To address this, we introduce the first large-scale aerial swarm flight dataset and online closed-loop verification platform tailored for LH tasks. The dataset comprises multimodal data collected collaboratively by 15 UAVs—4.32 million LiDAR frames and 12.96 million RGB frames—spanning 12 diverse scenarios, 720 trajectories, and 120 hours of flight time. The platform supports customizable simulation environments, sensor configurations, algorithm integration, and formation control modes, and features the first benchmark for wildlife conservation applications. We provide standardized evaluation benchmarks across nine state-of-the-art models, significantly advancing reproducible research and end-to-end system validation for autonomous swarm UAV navigation in LH settings.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Machine Learning: Large Multimodal Models (LMMs)
• Application Domains: Mobility, Driving & Flight

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Security and Privacy: Data transparency and provenance

Abstract

Swarm UAV autonomous flight for Long-Horizon (LH) tasks is crucial for advancing the low-altitude economy. However, existing methods focus only on specific basic tasks due to dataset limitations, failing in real-world deployment for LH tasks. LH tasks are not mere concatenations of basic tasks, requiring handling long-term dependencies, maintaining persistent states, and adapting to dynamic goal shifts. This paper presents U2UData-2, the first large-scale swarm UAV autonomous flight dataset for LH tasks and the first scalable swarm UAV data online collection and algorithm closed-loop verification platform. The dataset is captured by 15 UAVs in autonomous collaborative flights for LH tasks, comprising 12 scenes, 720 traces, 120 hours, 600 seconds per trajectory, 4.32M LiDAR frames, and 12.96M RGB frames. This dataset also includes brightness, temperature, humidity, smoke, and airflow values covering all flight routes. The platform supports the customization of simulators, UAVs, sensors, flight algorithms, formation modes, and LH tasks. Through a visual control window, this platform allows users to collect customized datasets through one-click deployment online and to verify algorithms by closed-loop simulation. U2UData-2 also introduces an LH task for wildlife conservation and provides comprehensive benchmarks with 9 SOTA models. U2UData-2 can be found at https://fengtt42.github.io/U2UData-2/.

Problem

• Addressing dataset limitations for long-horizon swarm UAV autonomous flight tasks
• Handling long-term dependencies and dynamic goal shifts in UAV operations
• Providing scalable data collection and algorithm verification for complex missions

Innovation

• Scalable online platform for swarm UAV data collection
• Large-scale dataset with multi-sensor environmental coverage
• Closed-loop simulation for algorithm verification and customization