AutomatumData/automatum-data-crossing

Automatum Data: T-Crossing Drone Dataset

Introduction

The Automatum Data T-Crossing Dataset contains high-precision movement data of traffic participants (cars, trucks, vans) extracted from drone recordings at T-shaped intersections in Bavaria, Germany. The data is captured from a bird's eye view and provides complete trajectories with velocities, accelerations, lane assignments, and object relationships.

This dataset directly competes with established benchmarks such as highD and NGSIM — offering superior data quality (JSON instead of CSV), standardized road geometry (OpenDRIVE XODR), and precise UTM world coordinate mapping.

Dataset at a Glance

Metric	Value
Scenario Type	T-Crossing / Intersection
Recordings	2
Total Duration	~30 minutes (0.49 hours)
Total Distance	108.8 km
Total Vehicles Tracked	683
Vehicle Types	623 Cars, 47 Trucks, 13 Vans
Max Trajectory Length	160.3 m
Coordinate System	UTM Zone 32U
FPS	29.97
License	CC BY-ND 4.0

Recording Overview

1. T-Crossing Gaimersheim Stadtweg

KPI	Value
Trajectories	299
Duration	650.7 s (~10.8 min)
Traffic Flow	1,654.3 veh/h
Traffic Density	40.2 veh/km
Avg. Speed	41.2 km/h
Max. Speed	109.4 km/h
Max. Acceleration	4.7 m/s²
Location	48.7882°N, 11.3855°E

2. T-Crossing St2214 Dünzlau Umgehung

KPI	Value
Trajectories	384
Duration	1,125.6 s (~18.8 min)
Traffic Flow	1,228.1 veh/h
Traffic Density	22.1 veh/km
Avg. Speed	55.6 km/h
Max. Speed	110.4 km/h
Max. Acceleration	5.8 m/s²
Location	48.7762°N, 11.3196°E

Data Structure

Each recording folder contains:

recording_folder/
├── dynamicWorld.json      # Trajectories, velocities, accelerations, bounding boxes
├── staticWorld.xodr       # Road geometry in OpenDRIVE format
├── recording_name.html    # Interactive metadata overview (Bokeh)
└── img/
    ├── kpis.json          # Key performance indicators
    ├── *_map.jpg          # Aerial map view
    ├── *_trajectories.jpg # Trajectory visualization
    └── *_centerImg_thumb.jpg  # Center frame thumbnail

dynamicWorld.json

The core data file contains for each tracked vehicle:

• Position vectors: x_vec, y_vec — UTM coordinates over time
• Velocity vectors: vx_vec, vy_vec — in m/s
• Acceleration vectors: ax_vec, ay_vec — in m/s²
• Jerk vectors: jerk_x_vec, jerk_y_vec
• Heading: psi_vec — orientation angle
• Lane assignment: lane_id_vec, road_id_vec — linked to XODR
• Object dimensions: length, width
• Object relationships: object_relation_dict_list — front/behind/left/right neighbors
• Safety metrics: ttc_dict_vec (Time-to-Collision), tth_dict_vec (Time-to-Headway)
• Lane distances: distance_left_lane_marking, distance_right_lane_marking

staticWorld.xodr

OpenDRIVE 1.6 format file defining:

• Road network topology and geometry
• Lane definitions with widths and types
• Junction configurations
• Speed limits and road markings

Key Metrics Explained

Quick Start

Installation

pip install openautomatumdronedata

Load and Explore

from openautomatumdronedata.dataset import droneDataset
import os

# Point to one recording folder
path = os.path.abspath("T-Crossing--GaimersheimStadtweg_e2e6-e2e6f4bb-4668-4654-ac7e-bcd90c9df4c2")
dataset = droneDataset(path)

# Access dynamic world
dynWorld = dataset.dynWorld

print(f"UUID: {dynWorld.UUID}")
print(f"Duration: {dynWorld.maxTime:.1f} seconds")
print(f"Frames: {dynWorld.frame_count}")
print(f"Vehicles: {len(dynWorld)}")

# Get all vehicles visible at t=1.0s
objects = dynWorld.get_list_of_dynamic_objects_for_specific_time(1.0)
for obj in objects[:5]:
    print(f"  {obj.UUID} ({obj.type}) — x={obj.x_vec[0]:.1f}, y={obj.y_vec[0]:.1f}")

Using with Hugging Face

from huggingface_hub import snapshot_download

# Download the dataset
local_path = snapshot_download(
    repo_id="AutomatumData/automatum-data-crossing",
    repo_type="dataset"
)

# Then load with openautomatumdronedata
from openautomatumdronedata.dataset import droneDataset
import os

recordings = [d for d in os.listdir(local_path) 
              if os.path.isdir(os.path.join(local_path, d)) 
              and not d.startswith('.') 
              and d not in ['doc', 'example_scripts', 'website_information']]

for rec in recordings:
    dataset = droneDataset(os.path.join(local_path, rec))
    print(f"{rec}: {len(dataset.dynWorld)} vehicles, {dataset.dynWorld.maxTime:.0f}s")

Example Scripts

See the example_scripts/ folder for ready-to-use analysis scripts:

• 01_lane_changes.py — Analyze lane change behavior across all vehicles
• 02_heatmap_density.py — Generate traffic density heatmaps
• 03_high_acceleration.py — Detect high-acceleration events
• 04_export_objects.py — Export per-vehicle JSON files with surrounding object data

Comparison with Established Datasets

Feature	Automatum Data	highD	NGSIM
Data Format	JSON + OpenDRIVE XODR	CSV + XML	CSV
Road Geometry	OpenDRIVE 1.6 standard	Simple annotations	Basic annotations
Coordinate System	UTM world coordinates	Local coordinates	Local coordinates
Object Relationships	Built-in (TTC, TTH, distances)	Must compute	Must compute
Velocity Error	< 0.2% (validated)	< 10 cm positional	Known issues
Python Library	openautomatumdronedata	Custom scripts	Custom scripts
OpenSCENARIO	Available on request	No	No

Research Use & Extended Data Pool

These publicly available datasets are intended exclusively for research purposes.

This dataset is a small excerpt from the comprehensive Automatum Data Pool containing over 1,000 hours of processed drone video. For commercial use or access to further datasets, including OpenSCENARIO exports, please contact us via our website:

automatum-data.com

Citation

If you use this dataset in your research, please cite:

@inproceedings{spannaus2021automatum,
  title={AUTOMATUM DATA: Drone-based highway dataset for development and validation of automated driving software},
  author={Spannaus, Paul and Zechel, Peter and Lenz, Kilian},
  booktitle={IEEE Intelligent Vehicles Symposium (IV)},
  year={2021}
}

License

This dataset is licensed under Creative Commons Attribution-NoDerivatives 4.0 International (CC BY-ND 4.0).

Contact

• Website: automatum-data.com
• Email: info@automatum-data.com
• HuggingFace: AutomatumData
• Documentation: openautomatumdronedata.readthedocs.io