This dataset is used for the classification of traffic intersection regulations using (car) GPS trajectory data. The coverage is mostly a large part of the city of Hannover, Germany.

The overview of the GPS trajectory dataset is given in the below figure: https://data.uni-hannover.de/dataset/379ed322-a9ea-48f3-bc13-2f5ea3174470/resource/ecfbc059-9ca3-4cbc-8774-de091f0fbcd6/download/hannover_traj.png" alt="Overview">

When the trajectory dataset is combined with the related intersection ground-truth information (available at: https://doi.org/10.25835/cqg0x1el): https://data.uni-hannover.de/dataset/379ed322-a9ea-48f3-bc13-2f5ea3174470/resource/bc832922-aedd-4bac-86b3-5d4a753bfecc/download/hannover_rules_traj.png" alt="Combined">

Data Acquisition

The trajectory samples were recorded using an android smartphone while driving a car in and around the city of Hannover, Germany. The acquisition period was from December 2017 to March 2019 by only a single person. The recording of the trajectories has taken place without restrictions in order to reflect a normal behavior of everyday car journeys. The sampling rate is approximately 1 sample per second.

Related Publications:

• Zourlidou, S., Sester, M. and Hu, S. (2022): Recognition of Intersection Traffic Regulations From Crowdsourced Data. Preprints 2022, 2022070012. DOI: https://doi.org/10.20944/preprints202207.0012.v1

• Zourlidou, S., Golze, J. and Sester, M. (2022): Traffic Regulation Recognition using Crowd-Sensed GPS and Map Data: a Hybrid Approach, AGILE GIScience Ser., 3, 22, 2022. https://doi.org/10.5194/agile-giss-3-22-2022

• Cheng, H., Lei, H., Zourlidou, S., Sester, M. (2022): Traffic Control Recognition with an Attention Mechanism Using Speed-Profile and Satellite Imagery data. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLIII-B4-2022, S. 287–29. https://doi.org/10.5194/isprs-archives-XLIII-B4-2022-287-2022

• Wang, C., Zourlidou, S., Golze, J. and Sester, M. (2020): Trajectory analysis at intersections for traffic rule identification. Geo-spatial Information Science, 11(4):1-10. https://doi.org/10.1080/10095020.2020.1843374

• Cheng, H., Zourlidou, S. and Sester, M. (2020): Traffic Control Recognition with Speed-Profiles: A Deep Learning Approach. ISPRS Int. J. Geo-Inf. 2020, 9, 652. https://doi.org/10.3390/ijgi9110652

• Golze, J., Zourlidou, S. and Sester, M. (2020): Traffic Regulator Detection Using GPS Trajectories. KN J. Cartogr. Geogr. Inf. https://doi.org/10.1007/s42489-020-00048-x

• Zourlidou, S., Fischer, C. and Sester, M. (2019): Classification of street junctions according to traffic regulators. In: Kyriakidis, P., Hadjimitsis, D., Skarlatos, D. and Mansourian, A., (eds) 2019. Accepted short papers and posters from the 22nd AGILE conference on geo-information science. Cyprus University of Technology 17–20 June 2019, Limassol, Cyprus.

Related Datasets:

• Zourlidou, S., Golze, J. and Sester, M. (2022). Dataset: Traffic Regulator Ground-truth Information for the Chicago Trajectory Dataset. https://doi.org/10.25835/0vifyzqi

• Zourlidou, S., Golze, J. and Sester, M. (2022). Dataset: GPS Trajectory Dataset and Traffic Regulation Information of the Region of Edessa, Greece. https://doi.org/10.25835/v0mzwob3

• Zourlidou, S., Golze, J. and Sester, M. (2022). Dataset: Traffic Regulator Ground-truth Information of the City of Hannover, Germany. https://doi.org/10.25835/cqg0x1el

• Zourlidou, S., Golze, J. and Sester, M. (2020). Dataset: Speed profiles and GPS Trajectories for Traffic Rule Recognition (6 Junctions, Hannover, Germany). https://doi.org/10.25835/0043786

TRACE-A_Trajectory_Data is the kinematic trajectory data collected during the Transport and Atmospheric Chemistry near the Equator - Atlantic (TRACE-A) suborbital campaign. Data from the Two Photon - Laser Induced Fluorescence (TP-LIF) and Differential Absorption of CO, CH4, N2O Measurements (DACOM) instruments are featured in this collection. Data collection for this product is complete.The TRACE-A mission was a part of NASA’s Global Tropospheric Experiment (GTE) – an assemblage of missions conducted from 1983-2001 with various research goals and objectives. TRACE-A was conducted in the Atlantic from September 21 to October 24, 1992. TRACE-A had the objective of determining the cause and source of the high concentrations of ozone that accumulated over the Atlantic Ocean between southern Africa and South America from August to October. NASA partnered with the Brazilian Space Agency (INPE) to accomplish this goal. The NASA DC-8 aircraft and ozonesondes were utilized during TRACE-A to collect the necessary data. The DC-8 was equipped with 19 instruments. A few instruments on the DC-8 include the Differential Absorption Lidar (DIAL), the Laser-Induced Fluorescence, the O3-NO Ethylene/Forward Scattering Spectrometer, the Modified Licor, and the DACOM IR Laser Spectrometer. The DIAL was responsible for a variety of measurements, which include Nadir IR aerosols, Nadir UV aerosols, Zenith IR aerosols, Zenith VS aerosols, ozone, and ozone column. The Laser-Induced Fluorescence instrument collected measurements on NxOy in the atmosphere. Measurements of ozone were recorded by the O3-NO Ethylene/Forward Scattering Spectrometer while the Modified Licor recorded CO2. Finally, the DACOM IR Laser Spectrometer gathered an assortment of data points, including CO, O3, N2O, CH4, and CO2. Ozonesondes played a role in data collection for TRACE-A along with the DC-8 aircraft. The sondes were dropped from the DC-8 aircraft in order to gather data on ozone, temperature, and atmospheric pressure.

mlfoundations-cua-dev/agent-trajectory-data dataset hosted on Hugging Face and contributed by the HF Datasets community

The Chicago data used for FHWA's Estimation of Travel Time Distributions Along User-Defined Travel Paths (EOTTD) project consisted of simulated vehicle trajectories, conducted for actual weather conditions.

The main dataset is a 232 MB file of trajectory data (I395-final.csv) that contains position, speed, and acceleration data for non-automated passenger cars, trucks, buses, and automated vehicles on an expressway within an urban environment. Supporting files include an aerial reference image (I395_ref_image.png) and a list of polygon boundaries (I395_boundaries.csv) and associated images (I395_lane-1, I395_lane-2, …, I395_lane-6) stored in a folder titled “Annotation on Regions.zip” to map physical roadway segments to the numerical lane IDs referenced in the trajectory dataset. In the boundary file, columns “x1” to “x5” represent the horizontal pixel values in the reference image, with “x1” being the leftmost boundary line and “x5” being the rightmost boundary line, while the column "y" represents corresponding vertical pixel values. The origin point of the reference image is located at the top left corner. The dataset defines five lanes with five boundaries. Lane -6 corresponds to the area to the left of “x1”. Lane -5 corresponds to the area between “x1” and “x2”, and so forth to the rightmost lane, which is defined by the area to the right of “x5” (Lane -2). Lane -1 refers to vehicles that go onto the shoulder of the merging lane (Lane -2), which are manually separated by watching the videos.

This dataset was collected as part of the Third Generation Simulation Data (TGSIM): A Closer Look at the Impacts of Automated Driving Systems on Human Behavior project. During the project, six trajectory datasets capable of characterizing human-automated vehicle interactions under a diverse set of scenarios in highway and city environments were collected and processed. For more information, see the project report found here: https://rosap.ntl.bts.gov/view/dot/74647. This dataset, which was one of the six collected as part of the TGSIM project, contains data collected from six 4K cameras mounted on tripods, positioned on three overpasses along I-395 in Washington, D.C. The cameras captured distinct segments of the highway, and their combined overlapping and non-overlapping footage resulted in a continuous trajectory for the entire section covering 0.5 km. This section covers a major weaving/mandatory lane-changing between L'Enfant Plaza and 4th Street SW, with three lanes in the eastbound direction and a major on-ramp on the left side. In addition to the on-ramp, the section covers an off-ramp on the right side. The expressway includes one diverging lane at the beginning of the section on the right side and one merging lane in the middle of the section on the left side. For the purposes of data extraction, the shoulder of the merging lane is also considered a travel lane since some vehicles illegally use it as an extended on-ramp to pass other drivers (see I395_ref_image.png for details). The cameras captured continuous footage during the morning rush hour (8:30 AM-10:30 AM ET) on a sunny day. During this period, vehicles equipped with SAE Level 2 automation were deployed to travel through the designated section to capture the impact of SAE Level 2-equipped vehicles on adjacent vehicles and their behavior in congested areas, particularly in complex merging sections. These vehicles are indicated in the dataset.

As part of this dataset, the following files were provided:

• I395-final.csv contains the numerical data to be used for analysis that includes vehicle level trajectory data at every 0.1 second. Vehicle type, width, and length are provided with instantaneous location, speed, and acceleration data. All distance measurements (width, length, location) were converted from pixels to meters using the following conversion factor: 1 pixel = 0.3-meter conversion.
• I395_ref_image.png is the aerial reference image that defines the geographic region and the associated roadway segments.
• I395_boundaries.csv contains the coordinates that define the roadway segments (n=X). The columns "x1" to "x5" represent the horizontal pi

Goddard’s LiDAR, Hyperspectral, and Thermal Imagery (G-LiHT) mission is a portable, airborne imaging system that aims to simultaneously map the composition, structure, and function of terrestrial ecosystems. G-LiHT primarily focuses on a broad diversity of forest communities and ecoregions in North America, mapping aerial swaths over Conterminous United States (CONUS), Alaska, Puerto Rico, and Mexico.The purpose of G-LiHT’s Trajectory data product (GLTRAJECTORY) is to provide aircraft location and orientation to support and supplement other G-LiHT data products.GLTRAJECTORY data are processed as a Google Earth overlay Keyhole Markup Language (KML) file over the extent of an entire flight path. A low resolution browse is also provided to show the flight path.

The main dataset is a 304 MB file of trajectory data (I90_94_stationary_final.csv) that contains position, speed, and acceleration data for small and large automated (L2) vehicles and non-automated vehicles on a highway in an urban environment. Supporting files include aerial reference images for six distinct data collection “Runs” (I90_94_Stationary_Run_X_ref_image.png, where X equals 1, 2, 3, 4, 5, and 6). Associated centerline files are also provided for each “Run” (I-90-stationary-Run_X-geometry-with-ramps.csv). In each centerline file, x and y coordinates (in meters) marking each lane centerline are provided. The origin point of the reference image is located at the top left corner. Additionally, in each centerline file, an indicator variable is used for each lane to define the following types of road sections: 0=no ramp, 1=on-ramps, 2=off-ramps, and 3=weaving segments. The number attached to each column header is the numerical ID assigned for the specific lane (see “TGSIM – Centerline Data Dictionary – I90_94Stationary.csv” for more details). The dataset defines six northbound lanes using these centerline files. Twelve different numerical IDs are used to define the six northbound lanes (1, 2, 3, 4, 5, 6, 10, 11, 12, 13, 14, and 15) depending on the run. Images that map the lanes of interest to the numerical lane IDs referenced in the trajectory dataset are stored in the folder titled “Annotation on Regions.zip”. Lane IDs are provided in the reference images in red text for each data collection run (I90_94_Stationary_Run_X_ref_image_annotated.jpg, where X equals 1, 2, 3, 4, 5, and 6).

This dataset was collected as part of the Third Generation Simulation Data (TGSIM): A Closer Look at the Impacts of Automated Driving Systems on Human Behavior project. During the project, six trajectory datasets capable of characterizing human-automated vehicle interactions under a diverse set of scenarios in highway and city environments were collected and processed. For more information, see the project report found here: https://rosap.ntl.bts.gov/view/dot/74647. This dataset, which is one of the six collected as part of the TGSIM project, contains data collected using the fixed location aerial videography approach with one high-resolution 8K camera mounted on a helicopter hovering over a short segment of I-94 focusing on the merge and diverge points in Chicago, IL. The altitude of the helicopter (approximately 213 meters) enabled the camera to capture 1.3 km of highway driving and a major weaving section in each direction (where I-90 and I-94 diverge in the northbound direction and merge in the southbound direction). The segment has two off-ramps and two on-ramps in the northbound direction. All roads have 88 kph (55 mph) speed limits. The camera captured footage during the evening rush hour (4:00 PM-6:00 PM CT) on a cloudy day. During this period, two SAE Level 2 ADAS-equipped vehicles drove through the segment, entering the northbound direction upstream of the target section, exiting the target section on the right through I-94, and attempting to perform a total of three lane-changing maneuvers (if safe to do so). These vehicles are indicated in the dataset.

As part of this dataset, the following files were provided:

• I90_94_stationary_final.csv contains the numerical data to be used for analysis that includes vehicle level trajectory data at every 0.1 second. Vehicle type, width, and length are provided with instantaneous location, speed, and acceleration data. All distance measurements (width, length, location) were converted from pixels to meters using the following conversion factor: 1 pixel = 0.3-meter conversion.
• I90_94_Stationary_Run_X_ref_image.png are the aerial reference images that define the geographic region for each run X.
• I-90-stationary-Run_X-geometry-with-ramps.csv contain the coordinates that define the lane centerlines for each Run X. The "x" and "y" columns represent the horizontal and ve

This dataset contains Back trajectories for the R/V Discoverer. These trajectories were calculated with the NOAA HYSPLIT 3.2 code using the NMC Global Data Assimilation System Medium Range Forecast model (MRF). 9-day back trajectories are computed at 00Z and 12Z daily, originating at the ship with a source altitude of 100 m and an averaging period of 24 hrs. This dataset contains ASCII files which describe the latitude, longitude and pressure level of trajectory points. A companion dataset contains gif images of each of these trajectories.

Researchers for the Next Generation Simulation (NGSIM) program collected detailed vehicle trajectory data on southbound US 101 and Lankershim Boulevard in Los Angeles, CA, eastbound I-80 in Emeryville, CA and Peachtree Street in Atlanta, Georgia. Data was collected through a network of synchronized digital video cameras.NGVIDEO, a customized software application developed for the NGSIM program, transcribed the vehicle trajectory data from the video. This vehicle trajectory data provided the precise location of each vehicle within the study area every one-tenth of a second, resulting in detailed lane positions and locations relative to other vehicles. Click the "Show More" button below to find additional contextual data and metadata for this dataset.

Surface trajectory data from GTS

All methods are described in the paper "Generation of Synthetic Urban Vehicle Trajectories", IEEE BigData 2022.

The city pair data set consists of one full year of 3536 trajectories flying from Paris--Orly (LFPO) to Toulouse--Blagnac (LFBO) airports between January and December 2017. This data set has been requested based on a set of 28 callsigns commonly attributed to trajectories serving this route.The airspace data set consists of seven months of trajectories flying through the LFBBPT airspace of the French Bordeaux Area Control Centre (ACC) between January 1st and August 6th 2017. The data set is limited to 14,461 trajectories crossing the airspace during the time intervals when the sector was operationally deployed according to the Sector Configurations Plans (SCP), also known as opening schemes. The goal of using the SCP is for the traffic under analysis to be representative of operational situations with a level of workload deemed acceptable by controllers.The landing data set consists of 19,480 trajectories landing at Zurich airport (LSZH) between October 1st and November 30rd 2019. We relied on The OpenSky Network database to properly label trajectories landing at LSZH.

TRACE-P_Trajectory_P3B_Data is the trajectory data collected onboard the P-3B aircraft during the Transport and Chemical Evolution over the Pacific (TRACE-P) suborbital campaign. Data from the Chemical Ionization Mass Spectrometer (CIMS) and the Differential Absorption of CO, CH4, N2O Measurements (DACOM) instruments are featured in this collection. Data collection for this product is complete.The NASA TRACE-P mission was a part of NASA’s Global Tropospheric Experiment (GTE) – an assemblage of missions conducted from 1983-2001 with various research goals and objectives. TRACE-P was a multi-organizational campaign with NASA, the National Center for Atmospheric Research (NCAR), and several US universities. TRACE-P deployed its payloads in the Pacific between the months of March and April 2001 with the goal of studying the air chemistry emerging from Asia to the western Pacific. Along with this, TRACE-P had the objective studying the chemical evolution of the air as it moved away from Asia. In order to accomplish its goals, the NASA DC-8 aircraft and NASA P-3B aircraft were deployed, each equipped with various instrumentation. TRACE-P also relied on ground sites, and satellites to collect data. The DC-8 aircraft was equipped with 19 instruments in total while the P-3B boasted 21 total instruments. Some instruments on the DC-8 include the Nephelometer, the GCMS, the Nitric Oxide Chemiluminescence, the Differential Absorption Lidar (DIAL), and the Dual Channel Collectors and Fluorometers, HPLC. The Nephelometer was utilized to gather data on various wavelengths including aerosol scattering (450, 550, 700nm), aerosol absorption (565nm), equivalent BC mass, and air density ratio. The GCMS was responsible for capturing a multitude of compounds in the atmosphere, some of which include CH4, CH3CHO, CH3Br, CH3Cl, CHBr3, and C2H6O. DIAL was used for a variety of measurements, some of which include aerosol wavelength dependence (1064/587nm), IR aerosol scattering ratio (1064nm), tropopause heights and ozone columns, visible aerosol scattering ratio, composite tropospheric ozone cross-sections, and visible aerosol depolarization. Finally, the Dual Channel Collectors and Fluorometers, HPLC collected data on H2O2, CH3OOH, and CH2O in the atmosphere. The P-3B aircraft was equipped with various instruments for TRACE-P, some of which include the MSA/CIMS, the Non-dispersive IR Spectrometer, the PILS-Ion Chromatograph, and the Condensation particle counter and Pulse Height Analysis (PHA). The MSA/CIMS measured OH, H2SO4, MSA, and HNO3. The Non-dispersive IR Spectrometer took measurements on CO2 in the atmosphere. The PILS-Ion Chromatograph recorded measurements of compounds and elements in the atmosphere, including sodium, calcium, potassium, magnesium, chloride, NH4, NO3, and SO4. Finally, the Condensation particle counter and PHA was used to gather data on total UCN, UCN 3-8nm, and UCN 3-4nm. Along with the aircrafts, ground stations measured air quality from China along with C2H2, C2H6, CO, and HCN. Finally, satellites imagery was used to collect a multitude of data, some of the uses were to observe the history of lightning flashes, SeaWiFS cloud imagery, 8-day exposure to TOMS aerosols, and SeaWiFS aerosol optical thickness. The imagery was used to best aid in planning for the aircraft deployment.

TRACE-P_Trajectory_DC8_Data is the trajectory data collected onboard the DC-8 aircraft during the Transport and Chemical Evolution over the Pacific (TRACE-P) suborbital campaign. Data collection for this product is complete.The NASA TRACE-P mission was a part of NASA’s Global Tropospheric Experiment (GTE) – an assemblage of missions conducted from 1983-2001 with various research goals and objectives. TRACE-P was a multi-organizational campaign with NASA, the National Center for Atmospheric Research (NCAR), and several US universities. TRACE-P deployed its payloads in the Pacific between the months of March and April 2001 with the goal of studying the air chemistry emerging from Asia to the western Pacific. Along with this, TRACE-P had the objective studying the chemical evolution of the air as it moved away from Asia. In order to accomplish its goals, the NASA DC-8 aircraft and NASA P-3B aircraft were deployed, each equipped with various instrumentation. TRACE-P also relied on ground sites, and satellites to collect data. The DC-8 aircraft was equipped with 19 instruments in total while the P-3B boasted 21 total instruments. Some instruments on the DC-8 include the Nephelometer, the GCMS, the Nitric Oxide Chemiluminescence, the Differential Absorption Lidar (DIAL), and the Dual Channel Collectors and Fluorometers, HPLC. The Nephelometer was utilized to gather data on various wavelengths including aerosol scattering (450, 550, 700nm), aerosol absorption (565nm), equivalent BC mass, and air density ratio. The GCMS was responsible for capturing a multitude of compounds in the atmosphere, some of which include CH4, CH3CHO, CH3Br, CH3Cl, CHBr3, and C2H6O. DIAL was used for a variety of measurements, some of which include aerosol wavelength dependence (1064/587nm), IR aerosol scattering ratio (1064nm), tropopause heights and ozone columns, visible aerosol scattering ratio, composite tropospheric ozone cross-sections, and visible aerosol depolarization. Finally, the Dual Channel Collectors and Fluorometers, HPLC collected data on H2O2, CH3OOH, and CH2O in the atmosphere. The P-3B aircraft was equipped with various instruments for TRACE-P, some of which include the MSA/CIMS, the Non-dispersive IR Spectrometer, the PILS-Ion Chromatograph, and the Condensation particle counter and Pulse Height Analysis (PHA). The MSA/CIMS measured OH, H2SO4, MSA, and HNO3. The Non-dispersive IR Spectrometer took measurements on CO2 in the atmosphere. The PILS-Ion Chromatograph recorded measurements of compounds and elements in the atmosphere, including sodium, calcium, potassium, magnesium, chloride, NH4, NO3, and SO4. Finally, the Condensation particle counter and PHA was used to gather data on total UCN, UCN 3-8nm, and UCN 3-4nm. Along with the aircrafts, ground stations measured air quality from China along with C2H2, C2H6, CO, and HCN. Finally, satellites imagery was used to collect a multitude of data, some of the uses were to observe the history of lightning flashes, SeaWiFS cloud imagery, 8-day exposure to TOMS aerosols, and SeaWiFS aerosol optical thickness. The imagery was used to best aid in planning for the aircraft deployment.

The dataset utilized in this paper is derived from the link-level trajectory data released by Didi for the ACM SIGSPATIAL GISCUP 2021 competition. This dataset, which has been widely used in various studies, provides desensitized trajectory data for Shenzhen in August 2020, along with the corresponding road network topology data.

Overview

This dataset provides high-resolution, georeferenced vehicle trajectories collected via drone footage at three roundabouts located in the municipalities of Frick and Laufenburg, Canton of Aargau, Switzerland. The data were collected as part of a collaborative drone campaign organized by the Urban Transport Systems Laboratory (LUTS), EPFL, within the framework of NCCR Automation, in cooperation with the cantonal traffic planning department of Aargau. The collection took place on Monday, 23rd October 2023, during peak morning and afternoon hours, resulting in nearly 11 hours of 4K video data.

Dataset Composition

This dataset contains CSV files structured with consistent data fields representing georeferenced trajectories, vehicle types (car, bus, truck), and timestamps, capturing detailed vehicle movements within roundabout environments.

File Organization

File names follow the convention:

D{X}_{TP}{N}_{S}.csv

• D{X} — the drone identifier, where {X} is a number (e.g., 1, 2) indicating which drone captured the data.
  → Example: D1 = data collected by Drone 1.
• {TP}{N} — the time period and session number, where {TP} is either AM (morning) or PM (afternoon), and {N} is an integer indicating the session number.
  → Example: AM2 = second morning session.
• {S} — the site identifier, corresponding to one of the monitored sites:
  → F1 = Roundabout F1 (Frick)
  → F2 = Roundabout F2 (Frick)
  → L1 = Roundabout L1 (Laufenburg)

CSV File Structure

Each CSV file includes:

Data Collection and Processing

• Collection Method: Two drones flying at an altitude of 120 meters above ground level, capturing videos at 4K resolution (3840×2160 pixels) at 29.97 FPS.
• Locations:
  • Roundabout F1 (Frick): Intersection of Bahnhofstrasse and Hauptstrasse 3 (Urban)
  • Roundabout F2 (Frick): Intersection of Hauptstrasse 3 with Gänsacker and Stöcklimattstrasse (Urban)
  • Roundabout L1 (Laufenburg): Intersection at Hauptstrasse 7 near the German border (Rural)
• Data Processing: The detection, tracking, and trajectory stabilization were performed using the early version of the Geo-trax framework (v0.1.0), an advanced computer vision pipeline tailored for drone-captured traffic footage. The resulting trajectories are precisely represented in stabilized pixel coordinates, which are subsequently transformed into geographic coordinates (WGS84). This georeferencing process follows a procedure similar to that described in Espadaler-Clapés et al., 2023, and includes:
  • Identification and extraction of Ground Control Points (GCPs) in the first stabilized video frame using QGIS Georeferencer, linking pixel coordinates to UTM coordinates.
  • Linear regression modeling between stabilized pixel coordinates and corresponding UTM coordinates derived from GCPs to estimate transformation parameters.
  • Projection to WGS84, converting UTM coordinates into global geographic coordinates using a standard GIS transformation (EPSG:4326).

Dataset Statistics

Potential Applications

This dataset is well-suited for:

• Gap acceptance behavior studies at roundabouts (e.g., Pascual Anglès et al., 2025)
• Traffic flow analysis and modeling
• Safety assessments using surrogate safety measures (SSMs)
• Validation of traffic simulation models

Large go-around, also referred to as missed approach, data set. The data set is in support of the paper presented at the OpenSky Symposium on November the 10th.

If you use this data for a scientific publication, please consider citing our paper.

The data set contains landings from 176 (mostly) large airports from 44 different countries. The landings are labelled as performing a go-around (GA) or not. In total, the data set contains almost 9 million landings with more than 33000 GAs. The data was collected from OpenSky Network's historical data base for the year 2019. The published data set contains multiple files:

go_arounds_minimal.csv.gz

Compressed CSV containing the minimal data set. It contains a row for each landing and a minimal amount of information about the landing, and if it was a GA. The data is structured in the following way:

    Column name
    Type
    Description




    time
    date time
    UTC time of landing or first GA attempt


    icao24
    string
    Unique 24-bit (hexadecimal number) ICAO identifier of the aircraft concerned


    callsign
    string
    Aircraft identifier in air-ground communications


    airport
    string
    ICAO airport code where the aircraft is landing


    runway
    string
    Runway designator on which the aircraft landed


    has_ga
    string
    "True" if at least one GA was performed, otherwise "False"


    n_approaches
    integer
    Number of approaches identified for this flight


    n_rwy_approached
    integer
    Number of unique runways approached by this flight

The last two columns, n_approaches and n_rwy_approached, are useful to filter out training and calibration flight. These have usually a large number of n_approaches, so an easy way to exclude them is to filter by n_approaches > 2.

go_arounds_augmented.csv.gz

Compressed CSV containing the augmented data set. It contains a row for each landing and additional information about the landing, and if it was a GA. The data is structured in the following way:

    Column name
    Type
    Description




    time
    date time
    UTC time of landing or first GA attempt


    icao24
    string
    Unique 24-bit (hexadecimal number) ICAO identifier of the aircraft concerned


    callsign
    string
    Aircraft identifier in air-ground communications


    airport
    string
    ICAO airport code where the aircraft is landing


    runway
    string
    Runway designator on which the aircraft landed


    has_ga
    string
    "True" if at least one GA was performed, otherwise "False"


    n_approaches
    integer
    Number of approaches identified for this flight


    n_rwy_approached
    integer
    Number of unique runways approached by this flight


    registration
    string
    Aircraft registration


    typecode
    string
    Aircraft ICAO typecode


    icaoaircrafttype
    string
    ICAO aircraft type


    wtc
    string
    ICAO wake turbulence category


    glide_slope_angle
    float
    Angle of the ILS glide slope in degrees


    has_intersection

string

    Boolean that is true if the runway has an other runway intersecting it, otherwise false


    rwy_length
    float
    Length of the runway in kilometre


    airport_country
    string
    ISO Alpha-3 country code of the airport


    airport_region
    string
    Geographical region of the airport (either Europe, North America, South America, Asia, Africa, or Oceania)


    operator_country
    string
    ISO Alpha-3 country code of the operator


    operator_region
    string
    Geographical region of the operator of the aircraft (either Europe, North America, South America, Asia, Africa, or Oceania)


    wind_speed_knts
    integer
    METAR, surface wind speed in knots


    wind_dir_deg
    integer
    METAR, surface wind direction in degrees


    wind_gust_knts
    integer
    METAR, surface wind gust speed in knots


    visibility_m
    float
    METAR, visibility in m


    temperature_deg
    integer
    METAR, temperature in degrees Celsius


    press_sea_level_p
    float
    METAR, sea level pressure in hPa


    press_p
    float
    METAR, QNH in hPA


    weather_intensity
    list
    METAR, list of present weather codes: qualifier - intensity


    weather_precipitation
    list
    METAR, list of present weather codes: weather phenomena - precipitation


    weather_desc
    list
    METAR, list of present weather codes: qualifier - descriptor


    weather_obscuration
    list
    METAR, list of present weather codes: weather phenomena - obscuration


    weather_other
    list
    METAR, list of present weather codes: weather phenomena - other

This data set is augmented with data from various public data sources. Aircraft related data is mostly from the OpenSky Network's aircraft data base, the METAR information is from the Iowa State University, and the rest is mostly scraped from different web sites. If you need help with the METAR information, you can consult the WMO's Aerodrom Reports and Forecasts handbook.

go_arounds_agg.csv.gz

Compressed CSV containing the aggregated data set. It contains a row for each airport-runway, i.e. every runway at every airport for which data is available. The data is structured in the following way:

    Column name
    Type
    Description




    airport
    string
    ICAO airport code where the aircraft is landing


    runway
    string
    Runway designator on which the aircraft landed


    n_landings
    integer
    Total number of landings observed on this runway in 2019


    ga_rate
    float
    Go-around rate, per 1000 landings


    glide_slope_angle
    float
    Angle of the ILS glide slope in degrees


    has_intersection
    string
    Boolean that is true if the runway has an other runway intersecting it, otherwise false


    rwy_length
    float
    Length of the runway in kilometres


    airport_country
    string
    ISO Alpha-3 country code of the airport


    airport_region
    string
    Geographical region of the airport (either Europe, North America, South America, Asia, Africa, or Oceania)

This aggregated data set is used in the paper for the generalized linear regression model.

Downloading the trajectories

Users of this data set with access to OpenSky Network's Impala shell can download the historical trajectories from the historical data base with a few lines of Python code. For example, you want to get all the go-arounds of the 4th of January 2019 at London City Airport (EGLC). You can use the Traffic library for easy access to the database:

import datetime from tqdm.auto import tqdm import pandas as pd from traffic.data import opensky from traffic.core import Traffic

load minimum data set

df = pd.read_csv("go_arounds_minimal.csv.gz", low_memory=False) df["time"] = pd.to_datetime(df["time"])

select London City Airport, go-arounds, and 2019-01-04

airport = "EGLC" start = datetime.datetime(year=2019, month=1, day=4).replace( tzinfo=datetime.timezone.utc ) stop = datetime.datetime(year=2019, month=1, day=5).replace( tzinfo=datetime.timezone.utc )

df_selection = df.query("airport==@airport & has_ga & (@start <= time <= @stop)")

iterate over flights and pull the data from OpenSky Network

flights = [] delta_time = pd.Timedelta(minutes=10) for _, row in tqdm(df_selection.iterrows(), total=df_selection.shape[0]): # take at most 10 minutes before and 10 minutes after the landing or go-around start_time = row["time"] - delta_time stop_time = row["time"] + delta_time

# fetch the data from OpenSky Network
flights.append(
  opensky.history(
    start=start_time.strftime("%Y-%m-%d %H:%M:%S"),
    stop=stop_time.strftime("%Y-%m-%d %H:%M:%S"),
    callsign=row["callsign"],
    return_flight=True,
  )
)

The flights can be converted into a Traffic object

Traffic.from_flights(flights)

Additional files

Additional files are available to check the quality of the classification into GA/not GA and the selection of the landing runway. These are:

validation_table.xlsx: This Excel sheet was manually completed during the review of the samples for each runway in the data set. It provides an estimate of the false positive and false negative rate of the go-around classification. It also provides an estimate of the runway misclassification rate when the airport has two or more parallel runways. The columns with the headers highlighted in red were filled in manually, the rest is generated automatically.

validation_sample.zip: For each runway, 8 batches of 500 randomly selected trajectories (or as many as available, if fewer than 4000) classified as not having a GA and up to 8 batches of 10 random landings, classified as GA, are plotted. This allows the interested user to visually inspect a random sample of the landings and go-arounds easily.

This GPS trajectory dataset was collected in (Microsoft Research Asia) Geolife project by 182 users in a period of over five years (from April 2007 to August 2012). A GPS trajectory of this dataset is represented by a sequence of time-stamped points, each of which contains the information of latitude, longitude and altitude. This dataset contains 17,621 trajectories with a total distance of 1,292,951kilometers and a total duration of 50,176 hours. These trajectories were recorded by different GPS loggers and GPS-phones, and have a variety of sampling rates. 91.5 percent of the trajectories are logged in a dense representation, e.g. every 1~5seconds or every 5~10 meters per point.This project uploads part of the Geolife dataset. Papers using this data include the following：[1] Yu Zheng, Lizhu Zhang, Xing Xie. Mining interesting locations and travel sequences from GPS trajectories. In Proceedings of International conference on World Wild Web (2009), Madrid Spain. ACM Press: 791-800[2] Yu Zheng, Quannan Li, Yukun Chen, Xing Xie. Understanding Mobility Based on GPS Data. In Proceedings of ACM conference on Ubiquitous Computing (UbiComp 2008), Seoul, Korea. ACM Press: 2008:312-321.[3] Yu Zheng, Xing Xie, Wei Ying Ma, GeoLife: A Collaborative Social Networking Service among User, location and trajectory. Invited paper, in IEEE Data Engineering Bulletin. 2010,2(33): 32-40.

Particle trajectory data from MD simulations utilizing the TIP4P/2005 water model and the Madrid-2019 force field. The simulation systems contain 4000 water molecules and different number of solute ion group (NaCl, KCl or CaCl2) to achieve concentration ranging from 0M to about 5.0M. The simulations were commenced within an NVT ensemble for 1ns and were then carried out in an NPT ensemble for 11ns with a time step of 1fs at a temperature ranging from 223.15K to 373.15K and a pressure of 1bar. Particle trajectories of a total of 101 moments were recorded at a uniform interval of 0.1ns in the last 10ns simulation.

Data to be found in the data fileset, metadata for each experiment will also be produced and posted, see links.

This is intended to become a repository for all data analysed with CeTrAn, an open source centroid trajetory analysis tool written in R (third link). The data will be uploaded while analysed (automatically), making it open by default.

I am looking for people interested in developing the tool and the database structure. I am no specialist, not in R, not in database structures, but I so far managed to find help in the R community. Come and comment, your feedback is needed !!