AutomatumData/automatum-data-highway-area-pools

Automatum Data: Highway Area Pools (Multi-Drone)

Digital twin of the pooled A9 Offenbau scene. Vehicles move along the real fused trajectories. Terrain, orthophoto and buildings are official Bavarian open geodata.

Introduction

Longer trajectories by flying two drones side by side and fusing their tracks into one scene.

The Automatum Data Highway Area Pools contain high-precision movement data of traffic participants (cars, trucks, vans, buses, motorcycles) extracted from drone recordings on German Autobahn segments. Each scene was recorded by two drones at the same time, hovering over neighbouring sections of the motorway. Each video is processed on its own by the Automatum pipeline: image stabilisation, CNN detection, georeferencing, tracking, perspective correction and vehicle dynamics. The two recordings are then pooled: we estimate the time offset between them, associate the vehicles seen by both drones in the overlap area and merge them into one consistent trajectory set.

On average, pooling makes the observed road section 1.9× longer than the best single drone. Vehicles that are associated across both drones are followed for more than a kilometre. This applies to 86 % of the vehicles at A9 Offenbau and to 34–39 % at the other two scenes. With a single drone, almost no vehicle is observed over 1 km.

The release contains 3 pools and the 6 single-drone recordings they are built from, in the native Automatum format (dynamicWorld.json + OpenDRIVE staticWorld.xodr, loadable with openautomatumdronedata) and additionally as Parquet tables for the Hugging Face dataset viewer.

Dataset at a Glance

Metric	Value
Scenario Type	Highway area pools (two drones side by side, fused tracks)
Pools	3 (built from 6 single-drone recordings)
Locations	A9 Offenbau (service area), A6 Schwabach-West (junction), A9 Lenting (construction-site layout, both directions on one carriageway)
Total Duration	~33 minutes of pooled scenes (0.55 hours); 66 minutes of drone video
Total Distance	3,619.5 km (pooled vehicle-km)
Total Vehicles Tracked	3,951 in the pools (6,087 in the single-drone recordings)
Vehicle Types (pools)	3,565 car, 179 truck, 125 van, 53 carWithTrailer, 20 bus, 9 motorcycle
Max Trajectory Length	1,307.6 m
Observed Road Length (pooled)	1,170 – 1,278 m per scene (single drone: 596 – 666 m)
Coordinate System	UTM Zone 32U (EPSG:32632)
FPS	29.97
Height	z_m (road surface, DHHN2016) in the pooled Parquet tables
License	CC BY-NC 4.0

KPI Comparison

KPIs per item, computed with the same script as for all Automatum datasets (img/kpis.json in every folder).

Item	Type	Trajectories	Duration [s]	Flow [veh/h]	Density [veh/km]	Avg. Traj. Length [m]	Avg. Speed [km/h]	Max. Speed [km/h]	Max. Acc. [m/s²]
Pool-Highway-A9-OffenbauRastanlage_c025-c02598c7-fa60-4302-8cee-12a05f4e0db6	pool	1,127	661.0	6,138.1	48.5	1,178.9	126.5	246.4	5.4
Highway-A9-OffenbauRastanlage_01_599d-599d5d23-a552-4067-84de-452df3de517f	drone 1	1,060	660.7	5,776.0	46.3	645.0	124.7	234.5	8.9
Highway-A9-OffenbauRastanlage_02_4d76-4d76cd4f-9594-40ea-8bfb-18c83a28eca0	drone 2	1,065	660.7	5,803.3	45.2	631.0	128.4	247.0	7.5
Pool-Highway-A6-SchwabachWest_b3c3-b3c33b28-c26b-44b7-a9ca-26cbfd9f02cc	pool	1,103	660.9	6,008.5	74.0	785.4	81.2	180.1	8.8
Highway-A6-SchwabachWest_01_9b40-9b402ea6-ddb2-42f0-afb2-ae40876adeb4	drone 1	709	660.7	3,863.4	50.0	535.2	77.3	137.7	8.3
Highway-A6-SchwabachWest_02_afbd-afbdc922-a6dc-493c-b24e-d8a07212fd94	drone 2	917	660.7	4,996.8	59.3	544.4	84.3	177.0	9.2
Pool-Highway-A9-LentingSolarpark_4d29-4d29e1dc-1237-4777-9a49-0189aea0b68b	pool	1,721	660.0	9,386.9	116.5	786.7	80.6	121.1	10.0
Highway-A9-LentingSolarpark_f043-f043acd1-97a7-4623-ab8a-61150a765d5e	drone 1	1,189	659.8	6,487.2	81.9	560.5	79.3	127.9	10.0
Highway-A9-LentingSolarpark_02_ba73-ba73af39-72c3-410a-9282-1caad9c274a5	drone 2	1,147	659.6	6,260.2	76.5	604.9	81.9	121.2	10.7

Avg. trajectory length is the mean distance driven per vehicle while it is observed. Max. speed and max. acceleration are the raw maxima over all samples and include single outliers (e.g. at the hand-over between the drones).

Repository Structure

highway-area-pools/
├── README.md                                   # This file
├── automatum.dataset.html                      # Interactive overview page (Bokeh), links to every item page
├── doc/                                        # Logo, illustrations, maps, technical diagrams
├── example_scripts/                            # Ready-to-use Python analysis scripts (01–04)
├── Sample_Data/                                # One pool unpacked for quick preview
│   └── Pool-Highway-A9-OffenbauRastanlage_c025-.../
│       ├── dynamicWorld.json
│       ├── Pool-Highway-A9-OffenbauRastanlage_c025-....html
│       └── img/
├── automatum_data_highway_area_pools.zip       # All 9 items (3 pools + 6 single-drone recordings)
├── data/<scene>/pooled|single/drone_1|single/drone_2/
│   ├── tracks.parquet, tracks_meta.parquet, scene.json   # Parquet tables (dataset viewer)
│   └── ../sync/time_sync_curve*.json           # time-offset search curves
├── metrics/pool_metrics.json                   # all numbers of the pooling tables
└── manifest.json                               # file list with sha256

Quick Preview: Browse Sample_Data/ to explore the data structure before downloading the full archive. The sample pool can be loaded directly with the openautomatumdronedata Python library. A9 Offenbau has no staticWorld.xodr yet; for a pool with road model use Pool-Highway-A6-SchwabachWest_... from the archive.

Items in the archive (folder names = original recording / pool names):

Scene	Pool	Drone 1	Drone 2
A9_OffenbauRastanlage	Pool-Highway-A9-OffenbauRastanlage_c025-c02598c7-fa60-4302-8cee-12a05f4e0db6	Highway-A9-OffenbauRastanlage_01_599d-599d5d23-a552-4067-84de-452df3de517f	Highway-A9-OffenbauRastanlage_02_4d76-4d76cd4f-9594-40ea-8bfb-18c83a28eca0
A6_SchwabachWest	Pool-Highway-A6-SchwabachWest_b3c3-b3c33b28-c26b-44b7-a9ca-26cbfd9f02cc	Highway-A6-SchwabachWest_01_9b40-9b402ea6-ddb2-42f0-afb2-ae40876adeb4	Highway-A6-SchwabachWest_02_afbd-afbdc922-a6dc-493c-b24e-d8a07212fd94
A9_LentingSolarpark	Pool-Highway-A9-LentingSolarpark_4d29-4d29e1dc-1237-4777-9a49-0189aea0b68b	Highway-A9-LentingSolarpark_f043-f043acd1-97a7-4623-ab8a-61150a765d5e	Highway-A9-LentingSolarpark_02_ba73-ba73af39-72c3-410a-9282-1caad9c274a5

Recording Overview

1. Highway A9 — Offenbau (service area)

Pool-Highway-A9-OffenbauRastanlage_c025-c02598c7-fa60-4302-8cee-12a05f4e0db6

KPI (pool)	Value
Trajectories	1,127
Duration	661.0 s (~11.0 min)
Traffic Flow	6,138.1 veh/h
Traffic Density	48.5 veh/km
Avg. Trajectory Length	1,178.9 m
Avg. Speed	126.5 km/h
Max. Speed	246.4 km/h
Max. Acceleration	5.4 m/s²
Location	49.1321°N, 11.2699°E
Static world (OpenDRIVE)	not yet (follows in a later version)

Single drones vs. pool

	Drone 1	Drone 2	Pool
Tracks	1,060	1,065	1,127
Observed road length	664 m	644 m	1,278 m
Median track length	675 m	652 m	1,298 m
90th percentile track length	681 m	656 m	1,307 m
Tracks > 1 km	0.1 %	0.0 %	86.4 %
Vehicle-km	690	677	1,339
Vehicle-hours	5.5	5.2	10.5
Vehicles associated across both drones			998
Estimated time offset drone 2 vs. drone 1			+0.12 s

Item pages: Pool-Highway-A9-OffenbauRastanlage_c025-c02598c7-fa60-4302-8cee-12a05f4e0db6 (in Sample_Data/); the single-drone item pages are inside the archive.

2. Highway A6 — Schwabach-West (junction)

Pool-Highway-A6-SchwabachWest_b3c3-b3c33b28-c26b-44b7-a9ca-26cbfd9f02cc

KPI (pool)	Value
Trajectories	1,103
Duration	660.9 s (~11.0 min)
Traffic Flow	6,008.5 veh/h
Traffic Density	74.0 veh/km
Avg. Trajectory Length	785.4 m
Avg. Speed	81.2 km/h
Max. Speed	180.1 km/h
Max. Acceleration	8.8 m/s²
Location	49.3092°N, 10.9961°E
Static world (OpenDRIVE)	yes (drone-1 section)

Single drones vs. pool

	Drone 1	Drone 2	Pool
Tracks	709	917	1,103
Observed road length	607 m	595 m	1,170 m
Median track length	615 m	602 m	642 m
90th percentile track length	641 m	628 m	1,218 m
Tracks > 1 km	0.0 %	0.0 %	38.8 %
Vehicle-km	392	516	894
Vehicle-hours	4.9	5.9	10.7
Vehicles associated across both drones			503
Estimated time offset drone 2 vs. drone 1			-0.58 s

Item pages: Pool-Highway-A6-SchwabachWest_b3c3-b3c33b28-c26b-44b7-a9ca-26cbfd9f02cc (inside the archive); the single-drone item pages are inside the archive.

3. Highway A9 — Lenting (construction-site layout, both directions on one carriageway)

Pool-Highway-A9-LentingSolarpark_4d29-4d29e1dc-1237-4777-9a49-0189aea0b68b

KPI (pool)	Value
Trajectories	1,721
Duration	660.0 s (~11.0 min)
Traffic Flow	9,386.9 veh/h
Traffic Density	116.5 veh/km
Avg. Trajectory Length	786.7 m
Avg. Speed	80.6 km/h
Max. Speed	121.1 km/h
Max. Acceleration	10.0 m/s²
Location	48.8184°N, 11.4697°E
Static world (OpenDRIVE)	yes (drone-1 section)

Single drones vs. pool

	Drone 1	Drone 2	Pool
Tracks	1,189	1,147	1,721
Observed road length	629 m	640 m	1,238 m
Median track length	615 m	636 m	640 m
90th percentile track length	650 m	677 m	1,289 m
Tracks > 1 km	0.0 %	0.0 %	34.5 %
Vehicle-km	681	713	1,386
Vehicle-hours	8.4	8.5	16.8
Estimated time offset drone 2 vs. drone 1			+0.20 s

Item pages: Pool-Highway-A9-LentingSolarpark_4d29-4d29e1dc-1237-4777-9a49-0189aea0b68b (inside the archive); the single-drone item pages are inside the archive.

Pooling summary and scene notes

scene	road	tracks (pooled)	road length, best single drone	road length, pooled	median track length, single	median track length, pooled	tracks > 1 km, single	tracks > 1 km, pooled	vehicle-km	estimated time offset
A6_SchwabachWest	A6, junction Schwabach-West	1103	607 m	1170 m	615 m	642 m	0.0 %	38.8 %	894	-0.58 s
A9_OffenbauRastanlage	A9, service area Offenbau	1127	664 m	1278 m	675 m	1298 m	0.1 %	86.4 %	1339	+0.12 s
A9_LentingSolarpark	A9, Lenting (solar park section)	1721	640 m	1238 m	636 m	640 m	0.0 %	34.5 %	1386	+0.20 s

In all three scenes the two drone footprints overlap by only about 25–30 m along the road.

• A9 Offenbau (service area): 94 % of the vehicles in the common time span are associated across the two drones (998 of 1065).
• A6 Schwabach-West (junction with ramps): 55 % are associated (503 of 915). The others remain separate tracks per drone.
• A9 Lenting. The recording is from November 2024 and shows a construction-site traffic layout: traffic in both directions runs on one carriageway (3 + 2 lanes). This is a rare scenario type. 54 % of the vehicles are associated (614 of 1146).

We are still investigating why the association rate differs so much between the scenes, given that the overlap is similar. A wider overlap is planned for future flights.

"Road length" is the 0.5–99.5 % extent of all vehicle positions along the main road axis. "Track length" is the distance driven per vehicle while it is observed.

How the drones are synchronised

The two videos are not frame-synchronised. The pooling step searches for the time offset Δt at which the vehicles inside the overlap area of both frames match best. For every candidate offset it records the mean position deviation and the number of vehicles that could be associated. data/<scene>/sync/time_sync_curve.json contains the whole curve, and every pool folder has the plot as img/<pool>_time_sync.png. In all three scenes the correct offset is a sharp, unambiguous peak: 330–420 associated vehicles within about ±0.2 s, compared with typically 20–60 elsewhere.

Two lessons from this release:

• Use the matched-vehicle count, not deviation alone. Minimising the deviation alone can lock onto spurious offsets that are supported by only a handful of vehicles. An earlier processing of A6 Schwabach-West had selected +11.8 s; the correct offset is −0.58 s.
• Do not trust metadata start times. The recording start times disagree with the true offset by 70 s in one scene, A9 Lenting, while the true offset is +0.19 s. A wide search with a coarse-to-fine step is the safer approach.

Data Structure

Each item folder (pool or single-drone recording) contains:

<item_name>/
├── dynamicWorld.json      # Trajectories, velocities, accelerations, jerk, curvature, object size
├── staticWorld.xodr       # Road geometry in OpenDRIVE format (not for A9 Offenbau)
├── staticWorld.rd5        # The same road model as RoadRunner scene (where available)
├── <item_name>.html       # Interactive metadata overview (Bokeh)
└── img/
    ├── kpis.json                      # Key performance indicators
    ├── <item_name>_map.jpg            # Orthophoto with the drone footprint(s)
    ├── <item_name>_trajectories.jpg   # Trajectory visualization
    ├── <item_name>_centerImg_thumb.jpg  # Center frame thumbnail (pools: drone 1 | drone 2)
    └── <item_name>_time_sync.png      # Pools only: time-offset search

dynamicWorld.json

Header: UTM-ReferencePoint, WGS84-ReferencePoint, RecordingName, UUID, Release, CalculationStepVersions, videoInfo (fps, frame count), object_counts, DrivenDistanceInMeter, MedianDrivenDistanceInMeter, contact and licence. For every tracked vehicle (objects):

• Identity and size: UUID, objType (car, van, truck, carWithTrailer, bus, motorcycle), length, width
• Time vector: time — seconds on the common time axis of the item
• Position vectors: x_vec, y_vec — metres relative to the UTM reference point
• Heading: psi_vec — orientation angle (unwrapped)
• Velocity vectors: vx_vec, vy_vec — along / across the vehicle's heading, in m/s
• Acceleration vectors: ax_vec, ay_vec — in m/s²
• Jerk vectors: jerk_x_vec, jerk_y_vec — in m/s³
• Curvature: curvature_vec — driven curvature in 1/m

Lane assignment, object relations, TTC/TTH and lane-marking distances are not part of this release (see the limitations below). openautomatumdronedata can compute lane assignment and object relations on demand from staticWorld.xodr (droneDataset.calculate_on_demand_values_for_all_objects()), where a road model is available.

staticWorld.xodr

OpenDRIVE 1.6 format file defining:

• Road network topology and geometry (incl. ramp connections)
• Lane definitions with widths and types
• Junction and ramp configurations
• Speed limits and road markings

The pooled scenes are expressed in the frame of drone 1, so their road model covers the drone-1 section.

Key metrics in other Automatum releases

The following metrics are part of other Automatum datasets and of the commercial data pool, but not of this release:

Parquet tables (data/)

data/<scene>/pooled/tracks.parquet         one row per vehicle and frame (29.97 Hz), fused scene
data/<scene>/pooled/tracks_meta.parquet    one row per vehicle (class, length, width, start/end time)
data/<scene>/pooled/scene.json             reference point (WGS84 + UTM), drone footprints (`frame_borders_utm`), object counts, versions
data/<scene>/single/drone_1|drone_2/...    the same tables for each drone on its own (for comparison)
data/<scene>/sync/time_sync_curve.json     time-offset search curve used for the pool (deviation + #matched vehicles per offset)
data/A9_LentingSolarpark/sync/time_sync_curve_wide.json   the ±20 s landscape for Lenting (final pool: ±2 s fine search)
metrics/pool_metrics.json                  all numbers of the pooling tables, plus histograms
manifest.json                              file list with sha256

column	unit	meaning
scene_id, track_id	–	scene and vehicle UUID
t_s, frame	s, –	time since scene start, frame index at 29.97 fps
x_vec, y_vec	m	position relative to the scene's UTM reference point (scene.json)
x_utm_m, y_utm_m	m	absolute position in WGS84 / UTM zone 32N (EPSG:32632)
z_m (pooled)	m	road-surface height above sea level (DHHN2016) at the vehicle position, from the official DGM1 terrain model and airborne LiDAR incl. bridge decks
psi_vec	rad	heading in the UTM frame, wrapped to (−π, π] in the Parquet tables (unwrapped in dynamicWorld.json)
vx_vec, vy_vec	m/s	velocity along / across the vehicle's heading (vehicle frame)
ax_vec, ay_vec	m/s²	acceleration along / across the vehicle's heading
jerk_x_vec, jerk_y_vec, curvature_vec	m/s³, 1/m	derived dynamics

Object classes: car, van, truck, carWithTrailer, bus, motorcycle.

Quick Start

Installation

pip install openautomatumdronedata setuptools

Load and Explore

from openautomatumdronedata.dataset import droneDataset
import os

# Point to one pool or recording folder
path = os.path.abspath("Sample_Data/Pool-Highway-A9-OffenbauRastanlage_c025-c02598c7-fa60-4302-8cee-12a05f4e0db6")
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}) — v={((obj.vx_vec[0]**2+obj.vy_vec[0]**2)**0.5)*3.6:.1f} km/h")

# Road model (None for A9 Offenbau, which has no staticWorld.xodr yet)
print(dataset.statWorld)

Using with Hugging Face

from huggingface_hub import snapshot_download, hf_hub_download
import zipfile

# Option 1: Download only the sample for a quick look
local_path = snapshot_download(
    repo_id="AutomatumData/automatum-data-highway-area-pools",
    repo_type="dataset",
    allow_patterns=["Sample_Data/**"]
)

# Option 2: Download the full archive (3 pools + 6 single-drone recordings)
archive = hf_hub_download(
    repo_id="AutomatumData/automatum-data-highway-area-pools",
    filename="automatum_data_highway_area_pools.zip",
    repo_type="dataset"
)
with zipfile.ZipFile(archive, "r") as z:
    z.extractall("automatum_data_highway_area_pools")

from openautomatumdronedata.dataset import droneDataset
dataset = droneDataset("automatum_data_highway_area_pools/Pool-Highway-A6-SchwabachWest_b3c3-b3c33b28-c26b-44b7-a9ca-26cbfd9f02cc")
print(f"Vehicles: {len(dataset.dynWorld)}")

# Option 3: Parquet tables (dataset viewer configs: pooled, pooled_meta, single_drone)
from datasets import load_dataset
pooled = load_dataset("AutomatumData/automatum-data-highway-area-pools", "pooled", split="train")
# or directly with pandas
import pandas as pd
df = pd.read_parquet("hf://datasets/AutomatumData/automatum-data-highway-area-pools/data/A9_OffenbauRastanlage/pooled/tracks.parquet")

Example Scripts

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

• 01_lane_changes.py — Analyze lane change behavior across all vehicles (lanes estimated from the lateral position)
• 02_heatmap_density.py — Generate traffic density heatmaps
• 03_high_acceleration.py — Detect high-acceleration events
• 04_pooling_gain.py — Compare a pool with its two single-drone recordings (road extent, track length distribution, share > 1 km)

cd example_scripts
python 04_pooling_gain.py ../Sample_Data/Pool-Highway-A9-OffenbauRastanlage_c025-c02598c7-fa60-4302-8cee-12a05f4e0db6

Processing versions and limitations (please read)

• Pipeline versions. The single-drone tracks come from the Automatum pipeline with image stabilisation 3.3 and detection 2.7. In this release the downstream steps were recomputed where the road model had changed. These are not the latest internal production versions (3.4 / 3.0). Each scene.json and dynamicWorld.json lists the exact CalculationStepVersions.
• Height. The x/y positions come from the production georeferencing, a plane fitted through ground control points. The pooled Parquet tracks also carry z_m, the height of the road surface at each sample. We took it from the Bavarian DGM1 terrain model and airborne LiDAR, with bridge decks extracted from the point cloud. The heights from the two drones agree to within a few centimetres. A fully terrain-aware re-projection of x/y improved the cross-drone consistency only at the scene with the strongest relief, so x/y are left unchanged in this version. Geodata: Bayerische Vermessungsverwaltung – www.geodaten.bayern.de, CC BY 4.0, data modified.
• Maps. staticWorld.xodr/.rd5 are the OpenDRIVE/RoadRunner models of each drone location. The pooled scene is expressed in the frame of drone 1, so its map covers the drone-1 section. Pool-wide maps are in preparation. A9 Offenbau has no map yet; one will follow in a later version.
• Licence change. Earlier Automatum releases on Hugging Face use CC BY-ND 4.0. This release uses CC BY-NC 4.0, so that researchers may modify the data, for example re-segment or fuse it.
• Ends of the frame. Tracks start and end at the outer border of the pooled frame. Vehicles entering or leaving at junctions and service areas are cut where they leave the observed area.
• Hand-over between drones. At the sample where a merged track passes from one drone to the other, the time step can deviate from 1/29.97 s by a few hundredths of a second (sub-frame time offset). If you derive speeds from positions, use t_s (Parquet) or time (JSON) rather than the frame index, or use the provided vx_vec.
• Lenting map. The OpenDRIVE/RoadRunner model for A9 Lenting shows the road layout before the construction works. It does not represent the construction-site layout of the recording.
• No lane-level relations. Lane assignment, object relations, TTC/TTH and lane-marking distances are not included.
• No raw video. No raw video and no personal data are included. Vehicles are represented only by their class, bounding-box size and motion.

Comparison with Established Datasets

Feature	Automatum Data Area Pools	highD	NGSIM
Data Format	JSON + OpenDRIVE XODR + Parquet	CSV + XML	CSV
Observed road length	~1.2–1.3 km (two drones, fused)	~420 m	~500–640 m
Coordinate System	UTM world coordinates (+ road height z_m)	Local coordinates	Local coordinates
Road Geometry	OpenDRIVE 1.6 standard (where available)	Simple annotations	Basic annotations
Python Library	openautomatumdronedata	Custom scripts	Custom scripts
OpenSCENARIO	Available on request	No	No

Research Use & Commercial Contact

These publicly available datasets are intended for non-commercial research and teaching.

This dataset is a small excerpt from the comprehensive Automatum Data Pool containing over 1,000 hours of processed drone video. For commercial use (OEM/Tier-1 validation, simulation, scenario databases), for custom recordings at your own locations, for larger multi-drone pools or OpenSCENARIO exports, please contact Automatum Data GmbH:

automatum-data.com/de · info@automatum-data.com

Citation

A paper describing the pooling method is in preparation. Until it is out, please cite:

@misc{automatum_highway_area_pools_1_0_0,
  title  = {AUTOMATUM Highway Area Pools: multi-drone fused highway trajectories},
  author = {Spannaus, Paul and Zechel, Peter and Lenz, Kilian},
  year   = {2026},
  publisher = {Hugging Face},
  howpublished = {\url{https://huggingface.co/datasets/AutomatumData/automatum-data-highway-area-pools}},
  note   = {Version 1.0.0}
}

@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-NonCommercial 4.0 International (CC BY-NC 4.0). Free for non-commercial research and teaching, with attribution. For commercial use, contact Automatum Data GmbH (see above).

Geodata Attribution

The height values (z_m), the orthophoto maps (img/*_map.jpg, doc/map_*.jpg, doc/pooled_orthophoto_*.jpg) and our 3D renderings (doc/illustration_pools.jpg, doc/icon_pools.jpg, doc/hero_*.jpg) use official open geodata: Datenquelle: Bayerische Vermessungsverwaltung – www.geodaten.bayern.de, Daten verändert (CC BY 4.0, https://creativecommons.org/licenses/by/4.0/). Products used: DOP20, DGM1, airborne laser point clouds, LoD2 buildings.

Contact

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