Loading...
Loading...
Self-driving perception, LiDAR/camera fusion, trajectory prediction, drone perception, robot manipulation
1,663 datasets
Indianapolis, Indiana, USA, is the focus of this dataset containing in situ airborne measurements of atmospheric carbon dioxide (CO2) from September 3, 2014. Data was collected during the morning commuter period using a NASA JPL-developed CO2 Laser Absorption Spectrometer (CO2LAS) onboard a DC-8 aircraft as part of the ASCENDS deployment to demonstrate IPDA lidar techniques. The measurements capture the CO2 plume downwind of the urban area, enabling emission rate estimations.
ExylosAi's dataset offers a third path for robot manipulation data, captured via consumer VR rather than expensive teleoperation farms or pure simulation. The dataset is structured for transfer-ready episodes and is compatible with the LeRobot framework. It was last updated on May 4, 2026.
IceBridge Sigma Space Lidar L0 Raw Time-of-Flight Data, Version 1 contains raw time-of-flight measurements for Antarctica and Greenland acquired by the Sigma Space Lidar. The data were collected by scientists working on the NSF and NERC-funded ICECAP project, with additional support from NASA Operation IceBridge. The dataset is available in BIN, HTML, and ISO formats.
Brisbane City Council provides data on designated recreational drone launch sites within its network of over 2,000 parks. The dataset identifies specific parks where remotely piloted aircraft can be launched and landed, supporting recreational use compliance. It is maintained by Brisbane City Council and was last updated in March 2026.
Featuring polygons representing natural waterway or channel structure locations, such as sediment ponds, litter traps, weirs, spillways, and drop structures. It was created by the Melbourne Water Corporation for asset management and maintenance purposes, with data maintained using Lidar survey data and 60-hectare catchment limits. The dataset includes associated details like descriptions and asset sections for As Constructed drawings.
Aggregating line objects delineating natural waterways managed by Melbourne Water, specifically those within catchments larger than 60 hectares. It includes attributes such as waterway name, unique asset identifier, and EPMS/asset section number for linking to associated drawings. The data is used for asset management, planning applications, hydrologic modeling, and flood impact analysis.
Featuring geographic points representing drain and waterway outlet locations within the Greater Melbourne region, including the Port Phillip Bay and Western Port coastlines. The data was created from the Vicmap Hydro streams dataset for catchments greater than 60 hectares and has been maintained using Lidar survey data. The author is the Melbourne Water Corporation.
13,386 newly identified geomorphic features, combined with 136 previously mapped features, detail the glacial landscape of northern Vancouver Island and surrounding marine areas. The dataset was created using high-resolution lidar DEMs, 3-meter satellite imagery, and ground truthing to map the retreat of the Cordilleran Ice Sheet. It provides a detailed record of complex multiphase ice retreat along coastal British Columbia.
Annual estimates of internal displacement in Iran, validated by the Internal Displacement Monitoring Centre. The data tracks two primary metrics: the year-end stock of Internally Displaced Persons (IDPs) and the flow of new displacement incidents, accounting for multiple displacements per person. This dataset is part of the Global Internal Displacement Database and was last updated in March 2026.
Autumn and winter multispectral UAV imagery from the Bosten Lake wetland in Xinjiang supports research on automated wetland land-cover mapping. The dataset includes code for an automated classification framework that uses unsupervised clustering to generate pseudo-labels for training a Random Forest classifier. This approach is designed for fine-scale mapping in spectrally complex environments, particularly when labeled samples are limited.
Building footprints in the Yukon Territory were extracted from LiDAR orthophotos using deep learning models. Manual corrections and regularization tools were applied to refine the geometry of these structures. The dataset is maintained by the Government of Yukon and was last updated in March 2026.
USIM is a large-scale underwater robot manipulation and navigation dataset collected in the Stonefish physics simulator. It contains 2,275 episodes across 20 tasks in 9 underwater scenarios, formatted in LeRobot v2.1 format with dual-camera video recordings. The dataset was created by Vincent2025hello and last updated in April 2026.
Scotland's LiDAR Phase 4 Digital Surface Model covers 17,945 square kilometres, though it does not provide full national coverage. The Scottish Government procured this dataset for public use in 2020 after it was initially captured by Fugro for Scottish Power Energy Networks from 2017 to 2019. The DSM was produced from point cloud data to monitor overhead power cable networks.
11,772 square kilometres of LiDAR point cloud data were captured for Scotland's power network and a council project. The Scottish Government procured this dataset for public use in 2019, with initial collection by Fugro for Scottish Power Energy Networks in 2015-2016. This dataset reflects LAS point cloud data with a density of 4 points per square metre.
17,945 square kilometres of Scottish terrain are represented in this Digital Terrain Model derived from LiDAR point cloud data. The Scottish Government procured this dataset for public use in 2020, based on aerial surveys conducted for Scottish Power Energy Networks from 2017 to 2019. This dataset reflects the bare-earth terrain model produced from the original point cloud data.
11,772 square kilometres of Scottish terrain are covered by this LiDAR-derived Digital Surface Model. The Scottish Government procured this dataset for public use in 2019, with data originally captured by Fugro for Scottish Power Energy Networks in 2015 and 2016. The dataset includes pilot flights from 2019 but does not provide full national coverage.
PrivateMap-Bench includes 15 real robot runs across 3 layout variants, producing 300 filtered artifacts, 2310 privacy rows, 5580 utility rows, and 1126 aggregate rows. It is a benchmark for studying the privacy–utility trade-off in indoor SLAM map sharing, evaluating floor-plan recovery, trajectory leakage, and navigation utility. The dataset was created by an anonymous author and last updated on May 7, 2026.
Lidar-derived raster data provides the first reflective surface, including treetops, rooftops, and ground features. The dataset is organized into 1 km by 1 km non-overlapping tiles compiled from multiple acquisition projects. It is produced by the Government of Ontario and was last updated in March 2026.
Bathymetry points contain water depth measurements collected across bodies of water in Ontario. Data originates from surveys conducted by the Government of Ontario using echo-sounders with GPS positioning and bathymetric LiDAR. The dataset was last updated in March 2026.
Ontario's bare-earth terrain is represented in this raster dataset derived from classified lidar point clouds. Data is organized into non-overlapping 1-km by 1-km tiles for download. The Government of Ontario compiled this dataset from multiple lidar acquisition projects, with specifications varying by project.