Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Tongtong Feng, Xin Wang, Feilin Han, Leping Zhang, Wenwu Zhu

TL;DR

ELH tasks demand long-horizon planning and persistent state maintenance in swarm UAVs, which existing datasets and tools fail to support. The authors present U2UData+ — a large-scale swarm UAV ELH dataset (15 UAVs, 12 scenes, 720 traces, 120 hours) and a scalable online collection plus closed-loop verification platform based on U2USim and UE5.2 — along with an ELH wildlife-conservation task and a 9-model benchmark suite. Contributions include a 3.62 TB dataset with rich multisensor data and environmental measurements, 12.96M RGB and depth frames, 4.32M LiDAR frames, and a wildlife-conservation ELH task, plus a platform for configurable simulators, sensors, formations, and tasks open-sourced for community use. This framework aims to bridge sim-to-real gaps, accelerate ELH swarm UAV development, and enable deployment in dynamic real-world environments.

Abstract

Swarm UAV autonomous flight for Embodied Long-Horizon (ELH) 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 ELH tasks. ELH tasks are not mere concatenations of basic tasks, requiring handling long-term dependencies, maintaining embodied persistent states, and adapting to dynamic goal shifts. This paper presents U2UData+, the first large-scale swarm UAV autonomous flight dataset for ELH 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 ELH 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 ELH 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+ also introduces an ELH task for wildlife conservation and provides comprehensive benchmarks with 9 SOTA models. U2UData+ can be found at https://fengtt42.github.io/U2UData-2/.

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

TL;DR

ELH tasks demand long-horizon planning and persistent state maintenance in swarm UAVs, which existing datasets and tools fail to support. The authors present U2UData+ — a large-scale swarm UAV ELH dataset (15 UAVs, 12 scenes, 720 traces, 120 hours) and a scalable online collection plus closed-loop verification platform based on U2USim and UE5.2 — along with an ELH wildlife-conservation task and a 9-model benchmark suite. Contributions include a 3.62 TB dataset with rich multisensor data and environmental measurements, 12.96M RGB and depth frames, 4.32M LiDAR frames, and a wildlife-conservation ELH task, plus a platform for configurable simulators, sensors, formations, and tasks open-sourced for community use. This framework aims to bridge sim-to-real gaps, accelerate ELH swarm UAV development, and enable deployment in dynamic real-world environments.

Abstract

Swarm UAV autonomous flight for Embodied Long-Horizon (ELH) 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 ELH tasks. ELH tasks are not mere concatenations of basic tasks, requiring handling long-term dependencies, maintaining embodied persistent states, and adapting to dynamic goal shifts. This paper presents U2UData+, the first large-scale swarm UAV autonomous flight dataset for ELH 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 ELH 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 ELH 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+ also introduces an ELH task for wildlife conservation and provides comprehensive benchmarks with 9 SOTA models. U2UData+ can be found at https://fengtt42.github.io/U2UData-2/.

Paper Structure

This paper contains 7 sections, 4 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. U2UData+ Platform
  4. U2UData+ Dataset
  5. U2UData+ Benchmark
  6. Conclusion
  7. Acknowledgments

Figures (4)

  • Figure 1: U2UData+ collects a large-scale swarm UAV autonomous flight dataset for ELH tasks. U2UData+ also bulid a scalable swarm UAV data online collection and algorithm closed-loop verification platform, supporting the customization of simulators, UAVs, sensors, flight algorithms, formation modes, and ELH tasks. Through a visual control window, U2UData+ allows users to collect customized datasets through one-click deployment online and to verify algorithms by closed-loop simulation.
  • Figure 2: The scalable UAVs data online collection and algorithm closed-loop verification platform. Wind throughout the map.
  • Figure 3: Visual control window. Users can collect customized datasets through one-click deployment online and verify algorithms by closed-loop simulation.
  • Figure 4: The visualization of the U2UData+ dataset. we select two swarm UAV collaboration clips and annotate them.