Airport Drone Detection Dataset: Training AI for Aviation's Toughest Challenge

The Problem Nobody Can Film

Here's a challenge for you: try getting permission to fly a drone near an active airport runway to collect training data. You won't. Flight regulations make it virtually impossible to capture real-world imagery of drones in airport environments—the exact scenarios where detection matters most.

Yet airports around the world face an escalating threat. Rogue drones have shut down major hubs like Gatwick and Newark, causing millions in losses and putting lives at risk. The technology to detect them exists, but the training data doesn't. Until now.

We created a high-fidelity synthetic dataset that solves this data collection paradox. No flight restrictions violated. No sensitive infrastructure filmed. Just the training data your Counter-UAS system actually needs.

Why Standard Drone Datasets Fail at Airports

Most public drone detection datasets show quadcopters against blue skies or simple outdoor backgrounds. That's nothing like what your model will encounter at an actual airport:

Our dataset addresses every one of these challenges with photorealistic synthetic imagery generated specifically for airport environments.

What's in the Dataset

Built for Real Detection Challenges

The difference between a demo and a deployed system is edge cases. We've specifically engineered this dataset to include the scenarios that break most drone detection models:

• Fog and rain: Reduced visibility conditions that challenge optical sensors
• Night and low-light: Golden hour, dusk, and nighttime operations
• Airport clutter: Hangars, terminals, ground equipment creating complex backgrounds
• Long-range targets: Small objects at perimeter defense distances
• Multiple drone types: Consumer quadcopters and potential threat variants

Dataset Structure

dataset/
  ├─ images/     # High-fidelity synthetic .jpg files
  └─ labels/     # YOLO format .txt annotations

Class Map:
ID | Class Name | Description
0  | drone      | Consumer quadcopters and rogue UAVs

YOLO Configuration

# Simuletic Airport Drone Configuration
path: /path/to/dataset
train: images
val: images

nc: 1
names: ['drone']

Use Cases

Download the Dataset

Privacy & Compliance

Citation

@dataset{simuletic_airport_drone_2025,
  author = {Simuletic Team},
  title = {Simuletic Airport Drone Threat & Safety Dataset},
  year = {2025},
  url = {https://simuletic.com}
}