Loading...
Loading...
Self-driving perception, LiDAR/camera fusion, trajectory prediction, drone perception, robot manipulation
2,004 datasets
L-band radar data from the Delta-X campaign provides water level change maps for the Atchafalaya and Terrebonne Basins in Louisiana. Three gridded products—temporalcoherence, waterlevelchange, and waterlevelchange_ramp—detail cumulative changes in centimeters at 30-minute intervals during spring and fall 2021 deployments. NASA's UAVSAR instrument on a Gulfstream-III aircraft collected these observations, which were validated against in-situ water level gauges.
NASA's UAVSAR L1 Single Look Complex stack products contain polarimetric L-band radar data collected during the Delta-X campaign over Louisiana's Atchafalaya Basin in spring and fall 2021. The dataset provides repeat-pass interferometric time series with 30-minute sampling intervals across HH, HV, VH, and VV polarizations, serving as the foundational data for deriving water level changes in wetlands. Data quality was validated against in-situ water level gauges deployed throughout the Mississippi River Delta floodplain.
Experimental simulation results comparing a proposed Vision-guided Multi-constraint RRT* (VM-RRT*) algorithm against the traditional RRT* algorithm for robotic arm path planning. The dataset, authored by Zhaopeng Yuan and last updated in April 2026, likely contains metrics such as planning time and end-effector motion parameters. The VM-RRT* algorithm achieved an average planning time of 3.27 seconds, approximately 20% faster than the traditional RRT*'s 4.07 seconds.
A 5.5 KB Excel file containing the standard Denavit-Hartenberg (DH) parameter table for an AUBO I5 robotic arm, uploaded by Zhaopeng Yuan in April 2026. The data is associated with research on a Vision-guided Multi-constraint RRT* algorithm for improving robotic grasping efficiency in chemical laboratory automation.
A 2017 raster elevation model covering over 60% of England at 1-meter spatial resolution. Produced by the Environment Agency, this Digital Surface Model (DSM) is derived from merged and re-sampled LIDAR archives, using the newest and best resolution data from repeat surveys. The dataset includes heights of objects like buildings and vegetation, has a vertical accuracy of +/-15cm RMSE, and is referenced to the Ordnance Survey Newlyn datum.
2017 LIDAR Composite DTM (Digital Terrain Model) is a raster elevation model covering approximately 75% of England at a 2-meter spatial resolution. Produced by the Environment Agency, it is derived from merged and re-sampled archival data, with the newest, best-resolution surveys used where available. The data has a vertical accuracy of +/-15cm RMSE and is presented in 5km grids referenced to Ordinance Survey Newlyn.
A 2-meter resolution raster elevation model covering approximately 75% of England, derived from airborne LIDAR surveys. Produced by the Environment Agency, the composite merges the best available data from a time-stamped archive and is updated annually. All LIDAR data has a vertical accuracy of +/-15cm RMSE and is referenced to the Ordnance Survey Newlyn datum.
England's terrain model derived from LIDAR data, covering over 60% of the country at a 1-meter spatial resolution. The Environment Agency produced this Digital Terrain Model in 2017 by processing the last-return LIDAR signal to remove surface objects like buildings and vegetation. Data is available in 5km grid tiles with a vertical accuracy of +/-15cm RMSE.
A 2026 vector map product provides fine-scale dendrometric characteristics for forest planning in Quebec. It includes predicted variables such as usable volume per hectare by species, basal area, and number of stems per hectare. The data is complementary to other forest inventories and covers territories with LiDAR acquisition in specific bioclimatic domains.
A Digital Terrain Model (DTM) derived from a 2021 pilot bathymetric LiDAR survey commissioned for Dundrum Bay and Carlingford Lough. The survey used a Rapid Airborne Multi-beam Mapping System (RAMMS) capable of approximately 25,000 range observations per second and penetration to three times visual water clarity. The dataset is provided by the Government Digital Service under the OGL-UK-3.0 license.
The NI 3D Coastal Survey commissioned a complete airborne LiDAR survey of the Northern Ireland coastline in 2021. The survey captured the intertidal area and extended approximately 200 meters landward of the high-water mark. This dataset is the Digital Surface Model derived from that LiDAR data.
A pilot bathymetric LiDAR survey commissioned in 2021 mapped the nearshore areas of Dundrum Bay and Carlingford Lough. The dataset is a Digital Terrain Model created from that survey, collected by the Government Digital Service under the OGL-UK-3.0 license. The survey used a Rapid Airborne Multi-beam Mapping System (RAMMS) capable of capturing high-resolution data to depths of three times the visual water clarity.
A Digital Terrain Model derived from a 2021 pilot bathymetric LiDAR survey. The survey mapped nearshore areas of Dundrum Bay and Carlingford Lough using a Rapid Airborne Multi-beam Mapping System (RAMMS). The dataset is published by the Government Digital Service under the OGL-UK-3.0 license.
Fifteen geotiff files contain processed bathymetry data from a 47-day survey conducted by RV Falkor from September 30 to November 17, 2020. The survey mapped submarine canyons, drowned reefs, and an underwater landslide in the Cape York Peninsula region and Coral Sea Marine Park using Kongsberg EM302 multibeam sonar. This dataset is published with the permission of Geoscience Australia.
IA-Bench (Interacted-Object Benchmark) provides human ground-truth annotations of interacted objects for robot manipulation subtasks. Each sample includes a full subtask video clip, gripper proprioception data aligned to frames, a language instruction, and pixel-coordinate bounding boxes for the initial and target object. The dataset is authored by irl-kit and was last updated on June 18, 2026.
July 2020 vegetation canopy data for Quebec City, derived from automated classification of World-View-3 and Pléiades satellite imagery with 31 cm resolution, supplemented by a 2017 Lidar survey. The dataset represents the ground projection of tree crowns as visible from the sky. It is provided by the Government and Municipalities of Québec under a CC-BY-4.0 license.
Edvinas Tiškus created a dataset evaluating changes in common reed (Phragmites australis) beds in Plateliai Lake. The data includes calculated NDWI and WAVI indices with land masking and validation polygons from Unmanned Aerial Vehicle imagery. The dataset is associated with a research paper on the PaperswithCode platform and is licensed as Open Access (green).
November 13 to December 7, 1991, this dataset contains vertical profiles of water vapor mixing ratio measured by a GSFC Raman Lidar during the second FIRE Cirrus intensive field observation in southeastern Kansas. Data consists of a series of one-minute and ten-minute averaged profiles with 75-meter vertical resolution from near ground level to 10.299 kilometers. The collection is part of the NASA-organized First ISCCP Regional Experiments aimed at improving cloud and radiation models.
From 1994 to 2024, this dataset compiles drone-related terrorist incidents worldwide by merging records from ACLED and the Global Terrorism Database. It focuses on the tactical use of Unmanned Aerial Systems by non-state actors and includes Python-based cleaning and merging logic. The dataset was authored by Dora Edelmann and is available under a CC-BY-4.0 license.
An archival dataset of question-and-answer content collected from the public website savollar.islom.uz. The answers reflect the views of specific scholars, primarily the late Shayx Muhammad Sodiq Muhammad Yusuf and associated scholars, as published on that site. The dataset was restructured for natural-language-processing research by author sukhrobnurali.