Loading...
Loading...
Self-driving perception, LiDAR/camera fusion, trajectory prediction, drone perception, robot manipulation
2,002 datasets
March 1998 and March 2005 LiDAR surveys provide 100-meter resolution GeoTIFFs of land-use, canopy height, and aboveground carbon estimates for La Selva Biological Station in Costa Rica. A supplementary look-up table relates modeled height changes to stand characteristics like age and carbon content. NASA produced this dataset to test the scale-dependency and accuracy of the Ecosystem Demography (ED) model for predicting vegetation dynamics and carbon flux.
OpenUAV-QA is a large-scale multiple-choice question-answering benchmark for UAV (drone) navigation decision-making, built upon the TravelUAV (OpenUAV) dataset. It transforms raw UAV flight trajectories into structured, text-polished 4-option QA pairs that test a multimodal model's ability to reason about path planning, action sequences, and spatial dynamics from first-person drone video footage. The dataset was created by choucsan and was last updated on 2026-07-12.
ANUGA 2D depth-integrated hydrodynamic model outputs and run-scripts for the Atchafalaya basin in Louisiana's Mississippi River Delta. The dataset contains over a month of simulation results from three field campaigns in fall 2016, spring 2021, and fall 2021. NASA produced this resource, which was calibrated using in-situ gauges and remote sensing data.
2021 data provides an updated multisource digital elevation model for the Atchafalaya and Terrebonne river basins in coastal Louisiana. It integrates sonar, bathymetric, and LiDAR data from the Delta-X and Pre-Delta-X campaigns, CPRA SWAMP, NOAA, and USGS. The dataset includes elevation values, a water/land mask, source flags, and weighting factor layers in cloud-optimized GeoTIFF format.
Fall 2017 aerosol and cloud measurements were collected over the United States by the High Spectral Resolution Lidar 2 (HSRL-2) instrument aboard a NASA ER-2 aircraft during the ACEPOL campaign. The data provide active remote sensing of aerosol height profiles, optical properties, and size distribution to improve understanding of aerosol effects on climate and air quality. Observations cover diverse scenes including urban, desert, forest, coastal ocean, and agricultural areas under clear, cloudy, polluted, and pristine conditions.
NASA provides 1 km resolution maps of woody vegetation cover and biomass for Sub-Saharan Africa from 2000 to 2004. The dataset integrates field measurements, Google Earth imagery, MODIS optical data, Q-SCAT microwave measurements, and spaceborne lidar. It was created using canopy cover observations, predicted woody cover, canopy height estimates, and tree allometry equations.
Fall 2016 lidar data from NASA's Pre-Delta-X campaign provides water surface elevation profiles for river channels between Wax Lake and the Gulf of Mexico. The dataset offers time-specific water levels in meters, calibrated with in-situ field measurements and referenced to NAVD 88 and WGS 84 geoids. It was produced by NASA using an Airborne Snow Observatory (ASO) lidar instrument.
Over 4500 emission factor bag measurements for CO2, CO, CH4, and N2O were collected using an unmanned aerial system. The dataset includes fuel parameters and fire severity proxies from 129 individual fires across savanna ecosystems in Brazil, Australia, Botswana, Zambia, South Africa, and Mozambique. Roland Vernooij of Vrije Universiteit Amsterdam compiled this data over six fire seasons between 2017 and 2022.
Airborne full waveform lidar and hyperspectral data in the VNIR bands collected by Flinders University's Airborne Research Australia aircraft. The record contains information on the Digital Elevation Model of the Alice Mulga site. Data was collected in 2014 and is hosted by the Terrestrial Ecosystem Research Network's Data Discovery portal.
A U.S. Geological Survey data release from January 15, 2026 defines dune crest, dune toe, and shoreline positions derived from lidar elevation surveys. The dataset includes calculated beach width and slope at 10-meter intervals alongshore for U.S. sandy coastlines. It was created by the Coastal Change Hazards Technical Capabilities and Applications project to assess vulnerability to storms and long-term shoreline change.
2012 elevation profile images of the Tumbarumba Wet Eucalypt Site obtained from airborne full waveform lidar and hyperspectral data. The data was collected by Flinders University's Airborne Research Australia (ARA) using a research aircraft. The record is provided by the Terrestrial Ecosystem Research Network's Data Discovery platform.
One million minutes of driving data collected from real user vehicles equipped with comma devices. The dataset includes road-camera video and full localization data for each segment. It was created by commaai and spans several hardware generations, including comma two, comma three, comma 3X, and comma four.
The western Pacific region during March-April 2001 is the focus of this dataset from NASA's Transport and Chemical Evolution over the Pacific (TRACE-P) campaign. It combines Differential Absorption Lidar (DIAL) remote sensing and in-situ instrument data from the DC-8 aircraft to measure trace gases, aerosols, ozone, and meteorological parameters. The dataset is part of the NASA Global Tropospheric Experiment (GTE) and represents a completed, multi-instrument atmospheric survey.
A vegetated river reach in the UK is represented by unfiltered point clouds. These directly georeferenced point clouds were created for PhD research, using UAV laser scanning (ULS) and structure-from-motion (SfM) derived from UAV multispectral (UAV-MS) data. The data is provided as supporting information by Chris Tomsett of the University of Southampton.
João Valente of Wageningen University & Research provides a dataset of orthophotos built from UAV images of a grassland field in Kleve, Germany. The images were captured at flight heights of 10, 15, and 30 meters. Annotations for detected Rumex obtusifolius (broad-leaved dock) plants are included.
Synchronized raw measurements from a coherent Doppler wind lidar, a reference flywheel, and an inclinometer used for calibration. The dataset includes multiple calibration runs performed at different reference speeds and laser focus settings. It was created by Anders Tegtmeier Pedersen of the Technical University of Denmark.
Humanoid Robot Manipulation Samples provide first-person manipulation sequences in real-world settings. The sample dataset, created by psdn-ai, is intended for reviewing household task structure, object interaction, and capture quality. It was last updated on July 7, 2026.
Processed MATLAB files for calibrating ABB's Micro-portable Greenhouse Gas Analyzer (MGGA) and a prototype version (pMGGA). The data supports the analysis in the journal article 'Testing the near-field Gaussian plume inversion flux quantification technique using unmanned aerial vehicle sampling' published in Atmospheric Measurement Techniques. Adil Shah from the University of Manchester authored the dataset.
Four flight experiments collected RGB and thermal infrared (TIR) imagery over two permafrost slopes in 2016 and 2017. The dataset, captured by UAV-mounted sensors, supports research on geomorphology and environmental change in the Qinghai-Tibet Engineering Corridor. It likely contains georeferenced images for analyzing slope stability and thermal properties.
Doppler lidar and ion spectrometer data from Hyytiälä, Finland, collected on May 1-2, 2017. Airborne measurement data from southern Finland was gathered using an instrumented Cessna 172 during several campaigns between 2011 and 2018. The dataset, compiled by Janne Lampilahti of the University of Helsinki, also includes new particle formation event classifications.