Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
124
datasets available to search
ShareScore release 0.9.0
Dataset results
124 results for “robotic dataset”
Dataset: First Trust Nasdaq Artificial Intelligence and Robotics ETF (ROBT) Stock Performance
This dataset provides historical stock market performance data for specific companies. It enables users to analyze and understand the past trends and fluctuations in stock prices over time. This information can be utilized for various purposes such as investment analysis, financial research, and market trend forecasting.
Dataset: ReWalk Robotics Ltd. (LFWD) Stock Performance
This dataset provides historical stock market performance data for specific companies. It enables users to analyze and understand the past trends and fluctuations in stock prices over time. This information can be utilized for various purposes such as investment analysis, financial research, and market trend forecasting.
Dataset: Nauticus Robotics, Inc. (KITT) Stock Performance
This dataset provides historical stock market performance data for specific companies. It enables users to analyze and understand the past trends and fluctuations in stock prices over time. This information can be utilized for various purposes such as investment analysis, financial research, and market trend forecasting.
Dataset: Nauticus Robotics, Inc. (KITTW) Stock Performance
This dataset provides historical stock market performance data for specific companies. It enables users to analyze and understand the past trends and fluctuations in stock prices over time. This information can be utilized for various purposes such as investment analysis, financial research, and market trend forecasting.
Dataset: Vaneck Robotics ETF (IBOT) Stock Performance
This dataset provides historical stock market performance data for specific companies. It enables users to analyze and understand the past trends and fluctuations in stock prices over time. This information can be utilized for various purposes such as investment analysis, financial research, and market trend forecasting.
Dataset for the article: Robotic Feet Modeled After Ungulates Improve Locomotion on Soft Wet Grounds
<div> <div>This repository contains data for three different experiments presented in the paper:</div> <br> <div>(1) moose_feet (40 files): The moose leg experiments are labeled as ax_y.nc,</div> <div>where 'a' indicates attached digits and 'f' indicates free digits. The</div> <div>number 'x' is either 1 (front leg) or 2 (hind leg), and the number 'y'</div> <div>is an increment from 0 to 9 representing the 10 samples of each set.</div> <br> <div>(2) synthetic_feet (120 files): The synthetic feet experiments are labeled</div> <div>as lw_a_y.nc, where 'lw' (Low Water content) can be replaced by 'mw'</div> <div>(Medium Water content) or 'vw' (Vast Water content). The 'a' can be 'o'</div> <div>(Original Go1 foot), 'r' (Rigid extended foot), 'f' (Free digits anisotropic</div> <div>foot), or 'a' (Attached digits). Similar to (1), the last number is an increment from 0 to 9.</div> <br> <div>(3) Go1 (15 files): The locomotion experiments of the quadruped robot on the</div> <div>track are labeled as condition_y.nc, where 'condition' is either 'hard_ground'</div> <div>for experiments on hard ground, 'bioinspired_feet' for the locomotion of the</div> <div>quadruped on mud using bio-inspired anisotropic feet, or 'original_feet' for</div> <div>experiments where the robot used the original Go1 feet. The 'y' is an increment from 0 to 4.</div> <br> <div>The files for moose_feet and synthetic_feet contain timestamp (s), position (m), and force (N) data.</div> <div>The files for Go1 contain timestamp (s), position (rad), velocity (rad/s), torque (Nm) data for all 12 motors, and the distance traveled by the robot (m).</div> <br> <div>All files can be read using xarray datasets (https://docs.xarray.dev/en/stable/generated/xarray.Dataset.html).</div> </div>
A dataset for robotic outdoor visual navigation with multiple passages through trajectory segments
<p>The images were captured by a fisheye camera and a magnetic compass was used to acquire the orientation data. The datasets are split in two folders:<br> 1) LEARN: In order to learn a new place, the robot camera captures 15 images over a 360 degrees panorama. During this process, the robot stays still in order to avoid distortions in the representation of the place.<br> 2) EXPLO: When exploring the environment (i.e. the rest of the time), the robot only captures 7 images per panorama, for the purpose of faster place recognition. Images are captured while the robot is moving. Various exploration panoramas are recorded around the trajectory performed in the learning panoramas (see traj.pdf).<br> <br> The average distance between two learning panoramas is 0.93 +/- 0.03 meters<br> The average distance traveled during an exploration panoramas is 0.71 +/- 0.01 meters<br> <br> DATASET A<br> ---------<br> - 20 meters long<br> - 22 learning panoramas (i.e. sets of 15 images captured while robot is stopped)<br> - 5 exploration trajectories<br> - A_on_learned: 29 exploration panoramas (i.e. sets of 7 images captured while robot is moving)<br> - A_parallel: 29 exploration panoramas<br> - A_diagonal1: 28 exploration panoramas<br> - A_diagonal2: 30 exploration panoramas<br> - A_diagonal3: 29 exploration panoramas<br> <br> DATASET B<br> ---------<br> - 20 meters long<br> - 21 learning panoramas (i.e. sets of 15 images captured while robot is stopped)<br> - 4 exploration trajectories<br> - B_on_learned: 29 exploration panoramas (i.e. sets of 7 images captured while robot is moving)<br> - B_parallel: 29 exploration panoramas<br> - B_diagonal1: 29 exploration panoramas<br> - B_diagonal2: 29 exploration panoramas<br> <br> DATASET C<br> ---------<br> - 23.1 meters long<br> - 25 learning panoramas (i.e. sets of 15 images captured while robot is stopped)<br> - 2 exploration trajectories<br> - C_on_learned: 34 exploration panoramas (i.e. sets of 7 images captured while robot is moving)<br> - C_parallel: 34 exploration panoramas<br> <br> <br> <br> PANO_INFO FILE STRUCTURE<br> ------------------------<br> Every folder containing images also contains an info file, named either learn_pano_info.SAVE or explo_pano_info.SAVE. Each line corresponds to an image. The structures is the following:<br> - column 1: id = image_id + 1<br> - column 2: azimuth of the center of the image in degrees/360 (value in [0,1])<br> - column 3: elevation of the center of the image. irrelevant in this database (equal to 0).<br> - column 4: type of panorama: equal to 1 if learning and to 0 if exploration.<br> - column 5: end of panorama: equal to 1 if it corresponds to the last image of a panorama.<br> <br> <br> REFERENCES<br> ----------<br> The dataset was used in the paper: Belkaid, M., Cuperlier, N., and Gaussier, P. Combining local and global visual information in context-based neurorobotic navigation. In Proceedings of the IEEE International Joint Conference on Neural Networks (IJCNN), pages 4947-4954, doi: 10.1109/IJCNN.2016.7727851, 2016.<br> <br> </p>
RSSI-based mobile robot localization datasets
<p>Experimental datasets used for an MSc project, titled "Outdoor Localization System for Mobile Robots Based on Radio-Frequency Signal Strength", for performing trajectory recovery of a mobile ground robot by data fusion of odometry, gyroscope and Received Signal Strength Indicator (RSSI), through Extended and Augmented-Extended Kalman Filter algorithms.</p> <p><br> "experiment_1" contains the robot's data in "rosbags" (i.e. ROS' compressed robot data), recorded in a parking lot with ground-truth from an RTK GPS. The measurements recorded were wheel odometry, IMU accelerations/angular velocities, RSSI from three receiver-transmitter pairs, regular GPS position and RTK GPS position.</p> <p>"experiment_2" contains the robot's data in "rosbags", recorded in the parking lot and in a second environment, a garden with tall trees and a nearby building. Both environments were recorded for later comparison, to see how the localization solution performed in a GPS-denied environment. All measurements were the same as in experiment 1, except for the exclusion of the common and RTK GPS.</p>
Human Grasp Dataset for Human Robot Handovers
<p>The dataset consists of 278.400 RGB images of size (299, 299).</p> <p>The images are sorted into the folders:</p> <ul> <li>Angles <ul> <li>30°, 45°, and 60°</li> </ul> </li> <li>Lights <ul> <li>From_Behind, From_Front, and Full_Lighting</li> </ul> </li> <li>Objects <ul> <li>Duplo_Block, Highlighter, Plastic_Pear, Table_Tennis_Racket, and Wood_Block</li> </ul> </li> <li>Persons <ul> <li>Person_1 to Person_10</li> </ul> </li> <li>Other <ul> <li>Default_Configuration, Clutter, and Other_Grasps_And_Interactions</li> </ul> </li> </ul> <p>Each folder contains 11.600 images that have a label in their file name. l=1 for grasp and l=0 for not grasp. </p>
Robot view — Supplementary dataset of experiment videos, IROS 2019 — Self-organized adaptive paths in multi-robot manufacturing: reconfigurable and pattern-independent fibre deployment
<p>This is a supplementary dataset of experiment videos of self-organized multi-robot fibre deployment. Each video is true speed and shows the full respective experiment. These videos show the <strong>robot view</strong> of each experiment.</p> <p><em>For a 2-minute summary video of these experiments, refer to:</em></p> <pre>https://doi.org/10.5281/zenodo.3357187</pre> <p>This supplementary dataset accompanies a conference paper prepared for IEEE IROS 2019.</p> <p>Using multi-robot systems for autonomous construction allows for parallelization and scalability. Swarm construction furthermore exploits robot interactions and collaboration, such that the robot swarm collectively constructs artifacts beyond what a single comparable robot could achieve. Here we present an alternative concept of swarm construction that is distinct because it uses continuous building material. Our approach is unique in its use of braiding techniques for construction. We deploy fibres that potentially allow for structures that are not possible with building blocks. To achieve maximal scalability we restrict ourselves to a decentralized approach. The main challenges are the local coordination of the robot teams, self-organized task allocation, and the dynamic reconfiguration of the braiding scheme at runtime. We successfully validate our approach in multi-robot experiments that show both braiding and branching of the braid. In addition, we show options for implementing an open system—that is robots can join and leave the braiding process on the fly.</p>
Fibre view — Supplementary dataset of experiment videos, IROS 2019 — Self-organized adaptive paths in multi-robot manufacturing: reconfigurable and pattern-independent fibre deployment
<p>This is a supplementary dataset of experiment videos of self-organized multi-robot fibre deployment. Each video is true speed and shows the full respective experiment. These videos show the <strong>fibre view</strong> of each experiment.</p> <p><em>For a 2-minute summary video of these experiments, refer to:</em></p> <pre>https://doi.org/10.5281/zenodo.3357187</pre> <p>This supplementary dataset accompanies a conference paper prepared for IEEE IROS 2019.</p> <p>Using multi-robot systems for autonomous construction allows for parallelization and scalability. Swarm construction furthermore exploits robot interactions and collaboration, such that the robot swarm collectively constructs artifacts beyond what a single comparable robot could achieve. Here we present an alternative concept of swarm construction that is distinct because it uses continuous building material. Our approach is unique in its use of braiding techniques for construction. We deploy fibres that potentially allow for structures that are not possible with building blocks. To achieve maximal scalability we restrict ourselves to a decentralized approach. The main challenges are the local coordination of the robot teams, self-organized task allocation, and the dynamic reconfiguration of the braiding scheme at runtime. We successfully validate our approach in multi-robot experiments that show both braiding and branching of the braid. In addition, we show options for implementing an open system—that is robots can join and leave the braiding process on the fly.</p>
RGB-Based Behavior Cloning Dataset for Surgical Robotics: 99,522 Episodes of Optimal Demonstrations
<h3><strong>Dataset Description</strong>:</h3> <p>This dataset contains 99,522 episodes of RGB-based state-action-reward expert demonstrations collected from a reaching task within a surgical robotics simulation environment, LapGym (Scheikl et al.). The data was generated using the LapGym ReachEnv, where a robotic grasper is tasked with reaching a specific point in 3D space. Each episode consists of a series of RGB images (64x64 pixels), corresponding actions, rewards, and terminal flags, designed for training behavior cloning and offline RL algorithms.</p> <p>This dataset was created for the paper "Assessing Behavior Cloning with RGB Inputs in Surgical Robotics Through Dataset Ablation". The expert demonstrations were collected using an optimal agent, where actions were computed based on the known locations of the grasper and the point of interest.</p> <p>The specific settings for the ReachEnv environment used to collect the dataset are as follows:</p> <ul> <li><strong>Environment</strong>: <code>ReachEnv</code></li> <li><strong>Observation Type</strong>: <code>RGB</code></li> <li><strong>Render Mode</strong>: <code>HUMAN</code></li> <li><strong>Action Type</strong>: <code>CONTINUOUS</code></li> <li><strong>Distance to Target Threshold</strong>: <code>0.01</code></li> <li><strong>Image Shape</strong>: <code>(64, 64)</code></li> <li><strong>Frame Skip</strong>: <code>1</code></li> <li><strong>Time Step</strong>: <code>0.1</code></li> <li><strong>Reward Amounts</strong>: <ul> <li><strong>Distance to Target</strong>: <code>0.0</code></li> <li><strong>Delta Distance to Target</strong>: <code>0.0</code></li> <li><strong>Successful Task</strong>: <code>100.0</code></li> <li><strong>Time Step Cost</strong>: <code>0.0</code></li> <li><strong>Workspace Violation</strong>: <code>0.0</code></li> </ul> </li> <li><strong>Sphere Radius</strong>: <code>0.020</code></li> </ul> <p>Key features of the dataset include:</p> <ul> <li><strong>RGB Inputs</strong>: Each episode includes 64x64 RGB frames representing the environment's visual state.</li> <li><strong>Optimal Demonstrations</strong>: All actions represent optimal behavior for completing the reach task.</li> <li><strong>Sparse Rewards</strong>: Rewards are only provided upon successful task completion, offering a challenging learning scenario.</li> <li><strong>Varied Episode Lengths</strong>: Episodes vary in length, depending on how quickly the task is completed.</li> </ul> <h3><strong>Applications</strong>:</h3> <p>This dataset is designed for research in:</p> <ul> <li>Behavior cloning with RGB image inputs.</li> <li>Data efficiency and sample efficiency in imitation learning.</li> <li>Offline reinforcement learning with visual inputs.</li> </ul> <h3><strong>Structure</strong>:</h3> <ul> <li><strong>Observations</strong>: Images stored as 64x64 RGB pixel arrays.</li> <li><strong>Actions</strong>: Continuous actions corresponding to the robotic grasper’s movements.</li> <li><strong>Rewards</strong>: Sparse rewards indicating task success.</li> <li><strong>Terminals</strong>: Terminal flags for task completion.</li> </ul> <h3><strong>How to Use</strong>:</h3> <p>This dataset can be used to train and evaluate offline models for robotic control tasks in conjunction with LapGym, particularly in the domain of surgical robotics. It is especially suited for behavior cloning experiments, offline reinforcement learning, and studies on data efficiency.</p> <h3><strong>Citation</strong>:</h3> <p>Please cite this dataset in any publications as:<br><em>Acs and Zhong (2024). RGB-Based Behavior Cloning Dataset for Surgical Robotics: 99,522 Episodes of Optimal Demonstrations. </em></p>
Spatial Room Impulse Response Dataset: A Robot's Journey Through Coupled Rooms of a Reverberant University Building
<p>This is a dataset of Spatial Room Impulse Responses obtained by a robot equipped with a microphone array.</p> <p>The measurements were conducted in a reverberant university building, the <em>Helmholtz</em> building at<em> Technische Universität Ilmenau</em> (coordinates: N50.6815788133375°, E10.939294371903342°). All the floors in the building are covered with bare stone tiles, the walls are not acoustically treated. Only the hallway has a suspended acoustic ceiling. The file "Pictures Overview.jpg" shows some impressions of the building. Note that the floorplan only shows parts of the building that were connected to the measurement area by open doors.</p> <p>The area covered by the robot is in a hallway on the top floor (2nd floor starting with ground floor) with two stairwells at both ends. To specifically study the behavior of coupled rooms and occluded sources, the sound sources were placed in adjacent sections of the building and on multiple floors. See the file "Measurement Overview.jpg" for an overview of the source positions and the receiver areas covered. Areas 2 and 3 were captured with a higher spatial resolution than area 1 to analyze the transition between the hallway and the staircases. The receiver positions form a uniform grid, the pitch between positions is shown in the following table. Due to time and technical constraints, only a maximum of 3 sources were used per run, so there are not all combinations of sources and receiver areas. Refer to the following table to see which source was active for which area and which zip file contains the according data:</p> <table> <tbody> <tr> <th>Filename</th> <th>Sources</th> <th>Receiver Area</th> <th>Receiver Positions [ct]</th> <th>Pitch [cm]</th> </tr> </tbody> <tbody> <tr> <td>Helmholtzbau_OG2_HM_HS.zip</td> <td>HM, HS</td> <td>Area 1</td> <td>143</td> <td>50</td> </tr> <tr> <td>Helmholtzbau_OG2_SML_SSL_SSU.zip</td> <td>SML, SSL, SSU</td> <td>Area 1</td> <td>154</td> <td>50</td> </tr> <tr> <td>Helmholtzbau_OG2_SMU_SML_HM.zip</td> <td>SMU, SML, HM</td> <td>Area 2</td> <td>88</td> <td>25</td> </tr> <tr> <td>Helmholtzbau_OG2_SSU_SSL_HS.zip</td> <td>SSU, SSL, HS</td> <td>Area 3</td> <td>92</td> <td>25</td> </tr> </tbody> </table> <p>As an example "Plot Reverberation Time.jpg" shows the reverberation times for all measured positions of Area 1 and 2 with speaker HM.</p>
Designed and generated Robot and Furniture datasets
<p>This dataset is a public data set designed to promote the application of artificial intelligence technology in reverse engineering. The dataset contains mesh objects with different parameters that are automatically generated from the manually designed CAD model database, and the labels are the specific parameter sizes corresponding to these objects. Researchers can read these objects and generate different variants for different tasks.</p>
ARIVVD: Aberystwyth Robot Infant Vision Video Dataset
<p>This is the Aberystwyth Robot Infant Vision Video Dataset, created for developmental robotics research at Aberystwyth University. This dataset was created for an internally funded research pilot project (“Babyvision: a robotic investigation into early development of colour constancy”). </p>
Dataset for Perspectives on Open Science and The Future of Scholarly Communication: Internet Trackers, Algorithmic Persuasion and Robotic Process Automation
<p>This data set was created between 01-04.2021, to study the current landscape of using web trackers in scholarly communication. The data set is part of an article (manuscript) that is intended to be published under the this title: Perspectives on Open Science and The Future of Scholarly Communication: Internet Trackers, Algorithmic Persuasion and Robotic Process Automation.</p>
A Physical Human-Robot Interaction Dataset - TacAct
<p>This dataset is the supplementary material of the IROS 2021 article "Organization and Understanding of a Tactile Information Dataset TacAct During Physical Human-Robot Interactions".<br> The dataset was collected by a flexible supercapacitor tactile sensor installed in an imitated mechanical arm device. In a 32 × 32 grid, the sensor data can be sampled at 100 Hz (100 frames per second). A total of 12 touch actions, namely, pull, squeeze, push, hold, grasp, poke, static drag, strong hit, soft slide, scratch, soft tap, and sliding drag, were recorded. A single-action collected from a subject consists of a 32×32×N matrix (where N is the number of frames or frame length). For the same action, all subjects were asked to use as many postures as possible to apply different forces to different positions of the sensor, and repeat 20 times with each hand. Except for strongly hit and soft tap (complete in an instant, acquisition time 2 s), the duration of each action is 2 s, and the time is 4 s altogether. The experiment was conducted on 50 subjects (36 males and 14 females,ranged from 22 to 36 years and 44 were right-handed) in total, each subject consists of 480 actions (12 actions × 40 repetitions ) in total, and the dataset collected 24,000 actions from the subjects.</p>
Robot@VirtualHome dataset
<p>The Robot@VirtualHome dataset is a raw collection of data from 30 virtual homes with different appearance obtained through the Robot@VirtualHome ecosystem. Each virtual house imitates a real house, keeping the same room layout and imitating real objects with virtual object models. The objectives of this dataset are: first, to be used as a testbed for diverse algorithms such as semantic mapping through the categorization of objects and/or rooms, active exploration of the environment, localization by appearance, or others where the data presented are of interest, and second to provide a basic example of the results that can be obtained through the Robot@VirtualHome ecosystem.</p> <p>The dataset consists of 113278 captures in 30 houses, with 236 rooms, 2569 objects and 4 different appearance conditions. Each data capture has stored an RGB image, a depth image, a semantic mask image, the measurements from a laser scanner and a log with information about the position at which the data was taken. In addition, for each house we have added the occupancy map obtained with the laser scanner and a log with the ground truth of all objects and rooms.</p> <p>Five raids have been carried out for each house: the first one, capturing data at the nodes of a grid using standard appearance, in the remaining four raids the data were taken by wandering around visiting all the rooms and using different appearance conditions.</p> <p>More detailed information is provided in the article.</p> <p>An API is available <a href="https://github.com/DavidFernandezChaves/RobotAtVirtualHome-Dataset-API">here </a>to facilitate access to the dataset data.</p>
Dataset for manuscript "Plants as inspiration for material‑based sensing and actuation in soft robots and machines"
<p>The dataset includes data for Figure 2 in the article "Plants as inspiration for material-based sensing and actuation in soft robots and machines<em>" MRS Bulletin</em> (2023). https://doi.org/10.1557/s43577-022-00470-8</p>
Robotic Monitoring of Alpine Screes: a Dataset from the EU Natura2000 habitat 8110 in the Italian Alps
<p>Data collected between the 19th and the 21rd of July 2022, in Valfurva, 23030 (SO), Italy, within the Stelvio National Park, located inside the Natura 2000 SPA IT2040044. The data acquisition has been conducted by a team composed of both robotic engineers and plant scientists. The platform used to collect the data is the ANYmal C quadrupedal robot.</p> <p> </p> <p>The dataset contains two different sets of data:</p> <p>1) typical and early warning species data - videos of seven different typical species of the habitat 8110 and one early warning species.</p> <p>2) monitoring mission data - video of the monitoring mission, robot status and point clouds, pictures and videos taken by the robot during the autonomous surveys.</p> <p> </p> <p>This dataset has a multidisciplinary scope and can be used by researchers in several fields. For instance, point clouds and information about the robot state could be used by robotic engineers to test or validate their own methods as well as benchmark the robot performance. On the other hand, plant videos and images recorded by the robot could be used by botanists to assess the quality of this information as well as the habitat's conditions, or by computer scientists interested in testing their AI algorithms for species detection and classification.</p>
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.