Swarm UAV Autonomous Flight Simulation for Long-Horizon Tasks

U2UData-2 is a large-scale simulation dataset for swarm UAV autonomous flight, specifically designed for Long-Horizon tasks. It was created by author fengtt42 and last updated on the Hugging Face platform in September 2025. The dataset addresses limitations in existing methods by modeling complex dependencies and dynamic goal shifts required for real-world deployment.

Use Cases

• Training reinforcement learning agents on long-horizon task sequences using simulated UAV state and action trajectories.
• Benchmarking swarm coordination algorithms against dynamic goal shift scenarios modeled in the simulation.
• Developing models for long-term dependency and persistent state maintenance using the provided task and environment features.

Strengths

• Described as the first large-scale swarm UAV dataset for Long-Horizon tasks.
• Models complex real-world requirements like long-term dependencies and dynamic goal shifts.

Limitations

• Specific row count, column details, and file size are unknown.
• Data is simulation-based, which may not fully capture real-world noise and physics.

Provenance

Source

Hugging Face dataset repository by author fengtt42.

Collection Method

Simulation data generated for swarm UAV autonomous flight research.

Freshness

Last updated on the platform in September 2025.