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”
PE-HRI-temporal: A Multimodal Temporal Dataset in a robot mediated Collaborative Educational Setting
<p><em><strong>Please note that this dataset corresponds to the training data used in "Social robots as skilled ignorant peers for supporting learning "[7]. This (second) version of the dataset additionally includes labels (PE score and cluster labels for each datapoint). </strong></em></p> <p> </p> <p>This data set consists of <strong>multi-modal temporal team behaviors as well as learning outcomes </strong>collected in the context of a robot mediated collaborative and constructivist learning activity called JUSThink [1,2]. The data set can be useful for those looking to explore evolution of log actions, speech behavior, affective states, and gaze patterns for students to model constructs such as engagement, motivation, collaboration, etc. in educational settings. </p> <p>In this data set, team level data is collected from 34 teams of two (68 children) where the children are aged between 9 and 12. There are two files: </p> <p><strong>PE-HRI_learning_and_performance.csv:</strong> This file consists of the <strong>team level performance and learning metrics</strong> which are defined below: </p> <ul> <li> <p><em>last_error:</em> This is the error of the last submitted solution. Note that if a team has found an optimal solution (error = 0) the game stops, therefore making last error = 0. This is a metric for performance in the task. </p> </li> <li> <p><em>T_LG_absolute:</em> It is a team-level learning outcome that we calculate by taking the average of the two individual absolute learning gains of the team members. The individual absolute gain is the difference between a participant’s post-test and pre-test score, divided by the maximum score that can be achieved (10), which grasps how much the participant learned of all the knowledge available.</p> </li> <li> <p><em>T_LG_relative:</em> It is a team-level learning outcome that we calculate by taking the average of the two individual relative learning gains of the team members. The individual relative gain is the difference between a participant’s post-test and pre-test score, divided by the difference between the maximum score that can be achieved and the pre-test score. This grasps how much the participant learned of the knowledge that he/she didn’t possess before the activity. </p> </li> <li> <p><em>T_LG_joint_abs: </em>It is a team-level learning outcome defined as the difference between the number of questions that both of the team members answer correctly in the post-test and in the pre-test, which grasps the amount of knowledge acquired together by the team members during the activity</p> </li> </ul> <p><strong>PE-HRI_behavioral_timeseries_w_labels.csv:</strong> In this file, for each team, the interaction of around 20-25 minutes is organized in windows of 10 seconds; hence, we have a total of 5048 windows of 10 seconds each. We report team level log actions, speech behavior, affective states, and gaze patterns for each window. More specifically, within each window, 26 features are generated in two ways: </p> <ol> <li>non-incremental</li> <li>incremental</li> </ol> <p>A non-incremental type would mean the value of a feature <em>in</em> that particular time window while an incremental type would mean the value of a feature <em>until</em> that particular time window. The incremental type is indicated by an "_inc" at the end of the feature name. Hence, in the end, within each window, we have 52 values: </p> <ul> <li> <p><em>T_add/(_inc): </em>The number of times a team added an edge on the map in that window/(until that window).</p> </li> <li> <p><em>T_remove/(_inc): </em>The number of times a team removed an edge from the map in that window/(until that window).</p> </li> <li> <p><em>T_ratio_add_rem/(_inc): </em>The ratio of addition of edges over deletion of edges by a team in that window/(until that window).</p> </li> <li> <p><em>T_action/(_inc):</em> The total number of actions taken by a team (add, delete, submit, presses on the screen) in that window/(until that window).</p> </li> <li> <p><em>T_hist/(_inc): </em>The number of times a team opened the sub-window with history of their previous solutions in that window/(until that window).</p> </li> <li> <p><em>T_help/(_inc): </em>The number of times a team opened the instructions manual in that window/(until that window). Please note that the robot initially gives all the instructions before the game-play while a video is played for demonstration of the functionality of the game. </p> </li> <li> <p><em>T1_T1_rem/(_inc): </em>The number of times either of the two members in the team followed the pattern consecutively: I add an edge, I then delete it in that window/(until that window).</p> </li> <li> <p><em>T1_T1_add/(_inc): </em>The number of times either of the two members in the team followed the pattern consecutively: I delete an edge, I add it back in that window/(until that window).</p> </li> <li> <p><em>T1_T2_rem/(_inc): </em>The number of times the members of the team followed the pattern consecutively: I add an edge, you then delete it in that window/(until that window).</p> </li> <li> <p><em>T1_T2_add/(_inc): </em>The number of times the members of the team followed the pattern consecutively: I delete an edge, you add it back in that window/(until that window).</p> </li> <li> <p><em>redundant_exist/(_inc): </em>The number of times the team had redundant edges in their map in that window/(until that window).</p> </li> <li> <p><em>positive_valence/(_inc): </em>The average value of positive valence for the team in that window/(until that window).</p> </li> <li> <p><em>negative_valence/(_inc): </em>The average value of negative valence for the team in that window/(until that window).</p> </li> <li> <p><em>difference_in_valence/(_inc): </em>The difference of the average value of positive and negative valence for the team in that window/(until that window).</p> </li> <li> <p><em>arousal/(_inc): </em>The average value of arousal for the team in that window/(until that window).</p> </li> <li> <p><em>gaze_at_partner/(_inc): </em>The average of the the two team member's gaze when looking at their partner in that window/(until that window). Each individual member's gaze is calculated as a percentage of time in that window/(until that window). </p> </li> <li> <p><em>gaze_at_robot/(_inc): </em>The average of the the two team member's gaze when looking at the robot in that window/(until that window). Each individual member's gaze is calculated as a percentage of time in that window/(until that window). </p> </li> <li> <p><em>gaze_other/(_inc): </em>The average of the the two team member's gaze when looking in the direction opposite to the robot in that window/(until that window). Each individual member's gaze is calculated as a percentage of time in that window/(until that window). </p> </li> <li> <p><em>gaze_at_screen_left/(_inc): </em>The average of the the two team member's gaze when looking at the left side of the screen in that window/(until that window). Each individual member's gaze is calculated as a percentage of time in that window/(until that window). </p> </li> <li> <p><em>gaze_at_screen_right/(_inc):</em> The average of the the two team member's gaze when looking at the right side of the screen in that window/(until that window). Each individual member's gaze is calculated as a percentage of time in that window/(until that window). </p> </li> <li> <p><em>T_speech_activity/(_inc): </em>The average of the two team member's speech activity in that window/(until that window). Each individual member's speech activity is calculated as a percentage of time that they are speaking in that window/(until that window). </p> </li> <li> <p><em>T_silence/(_inc): </em>The average of the two team member's silence in that window/(until that window). Each individual member's silence is calculated as a percentage of time in that window/(until that window). </p> </li> <li> <p><em>T_short_pauses/(_inc): </em>The average of the two team member's short pauses over their speech activity in that window/(until that window). Each individual member's short pause refers to a brief pause of 0.15 seconds and is calculated as a percentage of time in that window/(until that window). </p> </li> <li> <p><em>T_long_pauses/(_inc): </em>The average of the two team members long pauses over their speech activity in that window/(until that window). Each individual member's long pause refers to a pause of 1.5 seconds and is calculated as a percentage of time in that window/(until that window). </p> </li> <li> <p><em>T_overlap/(_inc): </em>The average percentage of time the speech of the team members overlaps in that window/(until that window).</p> </li> <li> <p><em>T_overlap_to_speech_ratio/(_inc): </em>The ratio of the speech overlap over the speech activity of the team in that window/(until that window).</p> </li> </ul> <p>Apart from these 52 values, within each window, we also indicate: </p> <ul> <li><em>team: </em>The team to which the window belongs to.</li> <li><em>time_in_secs:</em> Time in seconds until that window.</li> <li><em>window: </em>The window number.</li> <li><em>normalized_time: </em>The time when this window occurred with respect to the total duration of the task for a particular team. </li> <li>cluster_labels: The cluster number associated with each time window in reference to the productive and non-productive clusters found in [3]</li> <li>PE_score: The Productive Engagement score in each window</li> </ul> <p>Lastly, we briefly elaborate on how the features are operationalised. We extract log behaviors from the recorded rosbags while the behaviors related to both gaze and affective states are computed through the open source library OpenFace [6] that returns both facial actions units (AUs) as well as gaze angles. For voice activity detection (VAD), that classifies if a piece of audio is voiced or unvoiced, we made use of the python wrapper for the open source Google WebRTC VAD. The literature that inspired our log, audio and video features as well as the tools used to extract them are described in more detail in [3,4]. However, in those papers, we make use of only the aggregate version of this data [5].</p> <p><em><strong>Please note that this dataset corresponds to the training data used in [7]. This (second) version of the dataset additionally includes labels (PE score and cluster labels for each datapoint). </strong></em></p>
Visuo-motor dataset recorded from a micro-farming robot
<p>This is the accompanying dataset of the paper [1] describing algorithms for intrinsic motivation and episodic memory on the Sony LettuceThink microfarming robot.</p> <p>The LettuceThink microfarming robot developed by Sony Computer Science Laboratories consists of an aluminium frame with an X-Carve CNC machine mounted on it. The CNC machine is used to provide 3-axes movements to a depth camera (Sony DepthSense) mounted at the tip of the vertical z-axis (the end-effector camera). In the experiments presented in the paper, the end-effector camera is facing top-down and only two motors are used (x and y).</p> <p>A simulator of the LettuceThink robot has been developed to ease the testing of different configurations of the learning system. The simulator generates sensorimotor data from requested trajectories of the end-effector camera. Knowing the initial position of the CNC machine and the target position, the simulator linearly interpolates the trajectory and returns the intermediate positions of the camera together with the images captured from each specific position. The sensorimotor data returned by the simulator have been prerecorded by performing a full scan of the (x,y) plane of the CNC machine using a resolution of 5mm. This resulted in 24,964 images, each mapped to an (x,y) position of the CNC machine. The dataset published here contains these images.</p> <p>In particular, the dataset consists of a set of images, each named with the specific position of the 2 motors of the robot. A python script for generating visuo-motor trajectories (sequences of data consisting of [image, motor_x, motor_y]) from this dataset is available at the following github page: <a href="https://github.com/guidoschillaci/sonylettucethink_dataset">https://github.com/guidoschillaci/sonylettucethink_dataset</a></p> <p>Provided with the dataset is also a python script that allows to easily read the images and to generate trajectories (returning</p> <p>This work has been supported by the EU-H2020 ROMI Project and by the EU-H2020 Marie Sklodowska Curie project "Predictive Robots" (grant agreement no. 838861)References:</p> <p>[1] Schillaci, G., Villalpando, A. P., Hafner, V. V., Hanappe, P., Colliaux, D., & Wintz, T. (2020). Intrinsic Motivation and Episodic Memories for Robot Exploration of High-Dimensional Sensory Spaces. arXiv preprint arXiv:2001.01982.</p>
Robot@Home2, a robotic dataset of home environments
<p>The Robot-at-Home dataset (<strong>Robot@Home</strong>, paper <a href="http://mapir.uma.es/papersrepo/2017/2017-raul-IJRR-Robot_at_home_dataset.pdf">here</a>) is a collection of raw and processed data from five domestic settings compiled by a mobile robot equipped with 4 RGB-D cameras and a 2D laser scanner. Its main purpose is to serve as a testbed for semantic mapping algorithms through the categorization of objects and/or rooms.</p> <p>This dataset is unique in three aspects:</p> <ul> <li>The provided data were captured with a rig of 4 RGB-D sensors with an overall field of view of 180°H. and 58°V., and with a 2D laser scanner.</li> <li>It comprises diverse and numerous data: <em>sequences of RGB-D images and laser scans</em> from the rooms of five apartments (87,000+ observations were collected), <em>topological information</em> about the connectivity of these rooms, and <em>3D reconstructions</em> and <em>2D geometric maps</em> of the visited rooms.</li> <li>The provided ground truth is dense, including <em>per-point annotations</em> of the categories of the objects and rooms appearing in the reconstructed scenarios, and <em>per-pixel annotations</em> of each RGB-D image within the recorded sequences</li> </ul> <p>During the data collection, a total of 36 rooms were completely inspected, so the dataset is rich in contextual information of objects and rooms. This is a valuable feature, missing in most of the state-of-the-art datasets, which can be exploited by, for instance, semantic mapping systems that leverage relationships like <em>pillows are usually on beds</em> or <em>ovens are not in bathrooms</em>.</p> <p><strong>Robot@Home2</strong></p> <p><a href="https://github.com/goyoambrosio/RobotAtHome2">Robot@Home2</a>, is an enhanced version aimed at improving usability and functionality for developing and testing mobile robotics and computer vision algorithms. It consists of three main components. Firstly, a <a href="#bottom"><strong>relational database</strong></a> that states the contextual information and data links, compatible with Standard Query Language. Secondly,a <a href="https://pypi.org/project/robotathome/"><strong>Python package</strong></a> for managing the database, including downloading, querying, and interfacing functions. Finally, learning resources in the form of <a href="https://drive.google.com/drive/folders/1ENnxbKP5MJdlGl2Q93WTbIlofuy6Icxq"><strong>Jupyter notebooks</strong></a>, runnable locally or on the Google Colab platform, enabling users to explore the dataset without local installations. These freely available tools are expected to enhance the ease of exploiting the Robot@Home dataset and accelerate research in computer vision and robotics.</p> <p>If you use Robot@Home2, please cite the following <a href="https://www.sciencedirect.com/science/article/pii/S2352711023001863">paper</a>:</p> <p>Gregorio Ambrosio-Cestero, Jose-Raul Ruiz-Sarmiento, Javier Gonzalez-Jimenez, <strong>The Robot@Home2 dataset: A new release with improved usability tools</strong>, in <em>SoftwareX, Volume 23, 2023, 101490, ISSN 2352-7110</em>, <a href="https://doi.org/10.1016/j.softx.2023.101490">https://doi.org/10.1016/j.softx.2023.101490</a>.</p> <blockquote>@article{ambrosio2023robotathome2,<br>title = {The Robot@Home2 dataset: A new release with improved usability tools},<br>author = {Gregorio Ambrosio-Cestero and Jose-Raul Ruiz-Sarmiento and Javier Gonzalez-Jimenez},<br>journal = {SoftwareX},<br>volume = {23},<br>pages = {101490},<br>year = {2023},<br>issn = {2352-7110},<br>doi = {https://doi.org/10.1016/j.softx.2023.101490},<br>url = {https://www.sciencedirect.com/science/article/pii/S2352711023001863},<br>keywords = {Dataset, Mobile robotics, Relational database, Python, Jupyter, Google Colab}<br>}<br> </blockquote> <p><strong>Version history</strong><br><a href="../record/3901564">v1.0.1</a> Fixed minor bugs.<br><a href="../record/4495821">v1.0.2</a> Fixed some inconsistencies in some directory names. Fixes were necessary to automate the generation of the next version.<br><a href="../record/4499043">v2.0.0</a> SQL based dataset. Robot@Home v1.0.2 has been packed into a <a href="https://www.sqlite.org/index.html">sqlite</a> database along with RGB-D and scene files which have been assembled into a hierarchical structured directory free of redundancies. Path tables are also provided to reference files in both v1.0.2 and v2.0.0 directory hierarchies. This version has been automatically generated from version 1.0.2 through the <a href="https://github.com/goyoambrosio/RobotAtHome2">toolbox</a>.<br><a href="../record/4530453">v2.0.1</a> A forgotten foreign key pair have been added.<br><a href="../records/7811783">v.2.0.2</a> The views have been consolidated as tables which allows a considerable improvement in access time.<br><a href="../records/7811795">v.2.0.3</a> The previous version does not include the database. In this version the database has been uploaded.<br>v.2.1.0 Depth images have been updated to 16-bit. Additionally, both the RGB images and the depth images are oriented in the original camera format, i.e. landscape.</p>
PEDRo: an Event-based Dataset for Person Detection in Robotics
<p>This event-based dataset is specifically designed for person detection in robotics applications. </p>
THÖR-Magni (Demo Subset): a new multi-modal context-rich dataset of human-robot motion
<p>The Magni Human Motion Dataset provides high-quality tracking information from motion capture, eye-gaze trackers, and on-board robot sensors in a semantically rich environment. To induce natural behavior of recorded participants, we utilized loosely scripted task assignment, which induced participants to navigate through a dynamic laboratory environment in a natural and purposeful way. The dataset sets a high-quality standard as realistic and accurate data is enhanced with semantic information, enabling development of new algorithms that rely not only on tracking information but also on contextual cues of moving agents, static and dynamic environments.</p> <p> </p> <p>Link to dashboard that uses the data: https://magni-dash.streamlit.app/</p> <p><br> Here we publish a subset of the final dataset, to accompany the presentation at the 2023 IEEE International Conference on Robotics and Automation (ICRA)</p>
Dataset associated with article "Robots mediating interactions between animals for interspecies collective behaviors"
<p>This dataset contains results and analysis described in the study "Robots mediating interactions between animals for interspecies collective behaviors", Bonnet, F., Mills, R., Szopek, M., Schönwetter-Fuchs, S., Halloy, J., Bogdan, S., Correia, L., Mondada, F. and Schmickl, T. (2019), <em>Science Robotics</em>, <em>4</em>(28), doi: 10.1126/scirobotics.aau7897</p> <p>Contents: </p> <ul> <li>experimental data (logs from robotic systems, example videos)</li> <li>animal tracking analysis output</li> </ul> <p>See the readme and summary files contained within the archives for further details.</p>
Dataset of the scientific paper " Multimodal robotic system for upper-limb rehabilitation in physical environment" (Advances in Mechanical Engineering)
<p>There are eight files with the following information:<br> - pos_stateXX.bin, binary file with information of the end effector position of the robot device in meters along the three axis (X, Y, Z) during state XX of the experiment<br> - target_stateXX.bin, binary file with information of the target position for the robot device in meters along the three axis (X, Y, Z) during state XX of the experiment<br> - emg_channelXX.bin, binary file with information of channel 1 of the EMG sensor in mV during during the whole time of the experiment<br> - color_stateXX.bin, binary file with information of color filter information during state XX of the experiment. This information is the percentage of pixels with the correct color (yellow, cyan or magenta) inside the region of interest</p> <p> </p>
Passive Perching with Energy Storage for Winged Aerial Robots Dataset
<p>This dataset corresponds to the publication:</p> <p>"Passive Perching with Energy Storage for Winged Aerial Robots" W. Stewart, L. Guarino, Y. Piskarev, and D. Floreano. Advanced Intelligent Systems, <a href="http://doi.org/10.1002/aisy.202100150">http://doi.org/10.1002/aisy.202100150</a></p>
HRI30: An Action Recognition Dataset for Industrial Human-Robot Interaction
<p>A thorough analysis of the existing human action recognition datasets demonstrates that only a few HRI datasets are available that target real-world applications, all of which are adapted to home settings. Therefore, given the shortage of datasets in industrial tasks, we aim to provide the community with a dataset created in a laboratory setting that includes actions commonly performed within manufacturing and service industries. In addition, the proposed dataset meets the requirements of deep learning algorithms for the development of intelligent learning models for action recognition and imitation in HRI applications.</p>
Dataset for: Biohybrid superorganisms - on the design of a robotic system for thermal interactions with honeybee colonies
<p>Dataset containing electronic, mechical, firmware, and software design files associated with the article: </p> <p><br>"Biohybrid superorganisms - on the design of a robotic system for thermal interactions with honeybee colonies"<br>By R. Barmak, D. N. Hofstadler, M. Stefanec, L. Piotet, R. Cherfan, T. Schmickl, F. Mondada, and R. Mills. EPFL, Switzerland and Univeristy of Graz, Austria.<br>IEEE Access, 2024, Vol 12, pp 50849-50871.</p> <p>doi: 10.1109/ACCESS.2024.3385658</p> <p><a href="https://doi.org/10.1109/ACCESS.2024.3385658">https://doi.org/10.1109/ACCESS.2024.3385658</a></p> <p> </p> <table> <tbody> <tr> <td><strong>File name</strong></td> <td><strong>Description</strong></td> </tr> <tr> <td>1_hw_pcb_schematics_rev2.pdf</td> <td>Electrical schematic of the robotic frame</td> </tr> <tr> <td>2_hw_pcb_stackup_rev2.pdf </td> <td>Technical specifications for the robotic frame PCB manufacturing</td> </tr> <tr> <td>3_hw_pcb_bom_rev2.pdf </td> <td>Electronics Bill of Materials (BoM)</td> </tr> <tr> <td>4_hw_pcb_gerber_rev2.zip </td> <td>Robotic frame PCB manufacturing files (gerbers)</td> </tr> <tr> <td>5_hw_pcb_altium_rev2.zip</td> <td>Altium Designer project files</td> </tr> <tr> <td>6_hw_mechanical_rev2.zip</td> <td>DXF and STEP files of the mechanical structure of the robotic frame</td> </tr> <tr> <td>7_sw_firmware_rev2.zip</td> <td>Firmware source code and compiled binaries for STM32 microcontroller</td> </tr> <tr> <td>8_sw_handlers-1.0.1.zip</td> <td>Software for high-level interface to robot from a host device </td> </tr> </tbody> </table>
Hand gesture dataset based on sEMG data captured from the Technaid human-robot interaction system
<p>Two files with a dataset of five different/independent hand gestures are provided. The data were generated in a sEMG system with two bracelets (eight sEMG sensors and six sEMG sensors) worn in the right forearm of a human. The Technaid human-robot interaction system was used to captured the data. The file "datasetForSegmentation.mat" was used to train a classifier whose purpose is the execution of Segmentation process. On the other hand, the file "datasetForRecognition.mat" was used to train a classifier whose purpose is the execution of gesture Recognition process.</p> <p> </p>
Context-Aware 3D Object Anchoring for Mobile Robots Dataset
<p>This dataset accompanies the following publication:</p> <p>Günther, M.; Ruiz-Sarmiento, J. R.; Galindo, C.; González-Jiménez, J. & Hertzberg, J. <strong>Context-Aware 3D Object Anchoring for Mobile Robots.</strong> <em>Robot. Auton. Syst.</em>, 2018 (accepted)</p> <p>The dataset consists of 15 scenes inspected by a robot equipped with a RGB-D camera driving around a table and turning towards it from different locations. The table contained a number of objects in varying table settings. In total, the dataset contains 1387 seconds of observation and 144 unique objects from 9 categories:</p> <ul> <li>SugarPot</li> <li>MilkPot</li> <li>CoffeeJug</li> <li>MobilePhone</li> <li>Mug</li> <li>Dish</li> <li>Fork</li> <li>Knife</li> <li>Spoon</li> <li>TableSign</li> </ul> <p>Segmentation, tracking and local object recognition was run on the recorded sensor data, and its output (tracked objects and local recognition results) was added to the dataset. Since the objects were observed from multiple perspectives and tracking was lost while the robot was moving from one observation pose to another, the dataset contains more than one track ID for most objects (one for each subsequent observation of the object). Each track ID was manually labeled with the ground truth category of the object it represented. Additionally, all track IDs belonging to the same object were manually grouped together to allow evaluation of the anchoring process. Track IDs that did not correspond to any object on the table (but instead to objects on different tables, pieces of the table itself or other artifacts) were manually removed. In total, out of 432 track IDs, 410 (94.9 %) were associated with true objects, while 22 (5.1 %) were removed as artifacts.</p> <p><br> <strong>File contents</strong></p> <p>All data is provided as rosbags. The naming scheme is as follows:</p> <ul> <li>`*-sensordata.bag.bz2`: The raw sensor data from the robot and all transform data, including localization in a map.</li> <li>`*-perception.bag.bz2`: The object recognition results and ground truth information for the tracked objects.</li> <li>`scene??-pr2-*.bag.bz2`: 5 scenes that were recorded using the PR2 robot.</li> <li>`scene??-calvin-*.bag.bz2`: 10 scenes that were recorded using the Calvin robot.</li> </ul> <p>Both robots used an ASUS Xtion Pro Live as 3D camera.</p> <p>`race_vision_msgs.tar.bz2`: The custom messages used in the `-perception` rosbags, as a ROS Kinetic package.</p> <p><br> <strong>Videos</strong></p> <p>To get a first impression of the dataset, `scene10.mp4` and `scene19.mp4` show the corresponding scenes from the point of view of the robot's RGB camera.</p>
Arm gesture dataset based on IMU data captured from the Technaid human-robot interaction system
<p>Two files with a dataset of ten different/independent hand gestures are provided (seven static gestures and three dynamic gestures). The data were generated in a IMU system with five sensors worn in the right forearm, right arm, chest, left arm and left forearm of a human. The Technaid human-robot interaction system was used to captured the data. The file "datasetStaticGestures.mat" was used to train and test a classifier whose purpose is the recognition of static gestures. On the other hand, the file "datasetDynamicGestures.mat" was used to train and test a classifier whose purpose is the recognition of dynamic gestures. The latter file contains an extra class (gesture) which represents non-gestures.</p>
The DARRL dataset: Demonstrations for Action Recognition and Robot Learning
<p>The DARRL dataset (Demonstrations for Action Recognition and Robot Learning) is a collection of 760 RGB-D videos of humans performing various manipulation tasks. It is provided with object and action annotations (in the COCO format) for 30 of those videos; segmentation masks are also provided.</p> <p>It can also be used as a basis for learning from demonstrations for a robotic arm, for instance.</p> <p> </p> <p>This work is supported by Région Pays de la Loire.</p>
Will Biomimetic Robots Be Able to Change a Hivemind to Guide Honeybees' Ecosystem Services? (dataset)
<p>Simulation data for different simulation runs from the publication "Will Biomimetic Robots Be Able to Change a Hivemind to Guide Honeybees’ Ecosystem Services?"<br> <br> Dataset for figure 4: The Model replicates Seeleys Choice Experiment (1991).<br> (Shown in the folder: 'Model_Validation_Choice_Empirical','Model_Validation_Choice_Model')<br> Aditionally the accumulated energy [J] (net gain) gathered threw the foraging targets is displayed.<br> <br> Dataset for figure 5: The Model replicates Seeleys Cross Inhibition Experiment (2009).<br> (Shown in the folder: 'Model_Validation_Equilib_Empirical','SModel_Validation_Equilib_Model')<br> Aditionally the accumulated energy [J] (net gain) gathered threw the foraging targets is displayed.<br> <br> Dataset for figure 6: Model data of Seeley's choice experiment (1991) under more natural conditions.<br> The effect of the single parameters and the effect of all parameters together is shown.<br> (Shown in the folder: 'Natural_Foraging_Forager_Size','Natural_Foraging_crop_Load',<br> 'Natural_Foraging_CFS','Natural_Foraging_All_Conditions',)<br> <br> Dataset for figure 7: Model data of Seeley's Cross Inhibition Experiment under more natural conditions.<br> The effect of the single parameters and the effect of all parameters together is shown.<br> (Shown in the folder: 'Natural_Foraging_Equilib_Forager_Size','Natural_Foraging_Equilib_crop_Load',<br> 'Natural_Foraging_Equilib_CFS','Natural_Foraging_Equilib_All_Conditions',)<br> <br> Dataset for figure 8: The influence of waggle dancing robots on the foraging behaviour of honeybees with two different qualities of the foraging targets A and B.<br> After 14400 second (12:00) a pesticide is sprayed on the foraging target B under natural conditions.<br> The waggle dancing robot starts advertising foraging target B when time > 14400 seconds.<br> (The folder shows: 'robo_influence_bad_vs_bad','robo_influence_good_vs_bad','robo_influence_good_vs_good',)</p> <p>Dataset for figure 9: The influence of 10 waggle dancing robots on the foraging behaviour of honeybees with two different qualities of the foraging targets A and B<br> and varrying colony fill status (CFS).<br> After 14400 second (12:00) a pesticide is sprayed on the foraging target B under natural conditions.<br> The 10 waggle dancing robots start advertising foraging target B when time > 14400 seconds. The CFS is varried (0.1, 0.3, 0.5, 0.7, 0.9).<br> (The folder shows: 'robo_influence_bad_vs_bad','robo_influence_good_vs_bad','robo_influence_good_vs_good',)<br> <br> Dataset for figure 10: The effects at the end of the waggle dancing robots on the accumulated energy (through the trips), accumulated pesticides (through trips on foraging target<br> B, where after time>14400 pesticide was sprayed on the foraging target), pollination flights to foraging target A.<br> After 14400 second (12:00) a pesticide is sprayed on the foraging target B under natural conditions.<br> The waggle dancing robot starts advertising foraging target A when time > 14400 seconds.<br> (The folder shows: 'robo_influence_acc_energy','robo_influence_acc_pesticide','robo_influence_pollination')<br> <br> The datasets are contained in the zipped file folder Figures_data.zip.</p>
Hazards&Robots: A Dataset for Visual Anomaly Detection in Robotics
<p>This is the final version of our dataset; we further expand the Corridor scenario.</p> <p>This new version of Corridor includes 20 anomalies and the total frames are 324,408.</p> <p>In this version, we release feature embeddings extracted using a CLIP ViT-B/32 model.</p> <p>This dataset is part of a Data in Brief paper submission.</p> <p>For more information check https://github.com/idsia-robotics/hazard-detection</p> <p> </p>
Supplementary codes and datasets for "Wang tiles enable combinatorial design and robot-assisted manufacturing of modular mechanical metamaterials"
<p>This repository provides data and codes for manuscript “Wang tiles enable combinatorial design and robot-assisted manufacturing of modular mechanical metamaterials” by M. Doškář, M. Somr, R. Hlůžek, J. Havelka, J. Novák, and J. Zeman, published first as a preprint <a href="https://arxiv.org/abs/2305.09280">arXiv:2305.09280</a> at arXiv.org; see the actual description of the Zenodo entry for the latest reference.</p> <p>This repository contains:</p> <ol> <li>MATLAB and C++ source codes for combinatorial design and numerical analyses (folder <code>./numerics/</code>),</li> <li>experimental data (folder <code>./experiments/</code>),</li> <li>3D models of parts used in robotic-assembly (folder <code>./models/</code>),</li> <li>a control script for robotic assembly (folder <code>./robotics/</code>).</li> </ol> <p><strong>Numerics</strong></p> <p>All simulations were performed with an in-house MATLAB code, which extends the finite element toolbox for finite strain calculations accompanying the work of <a href="https://doi.org/10.1016/j.cma.2020.113333">van Bree, S. E. H. M., Rokoš, O., Peerlings, R. H. J., Doškář, M., & Geers, M. G. D. (2020). A Newton solver for micromorphic computational homogenization enabling multiscale buckling analysis of pattern-transforming metamaterials. Computer Methods in Applied Mechanics and Engineering, 372, 113333</a>. In particular, this snapshot corresponds to a cleaned-up version (excluding files unrelated to the publications) of commit <code>21cfc2e9</code>.</p> <p>The MATLAB codebase contains MEX files written in C++ to accelerate selected procedures. In order to run any code, these MEX files must be compiled first. We use CMake build automation, with the main <code>CMakeLists.txt</code> located in <code>./numerics/mex</code>.</p> <p>Combinatorial search was performed by the <code>RUN_modular_exploration.m</code> script; see definition of problems with the script. The results of the enumerations, stored in <code>./dat/exploration</code>, were analysed with <code>POST_modular_S_v3.m</code>, identifying layouts leading to the extreme (min/max) tilt angles.</p> <p>Comparison against experimental measurements was facilitated by a series of scripts <code>POST_DIC_{...}.m</code>. First, run <code>POST_DIC_step1_extract_points_in_mesh.m</code> to identify locations.mat. Next, post-process extensometer data with <code>POST_DIC_step2_merge_extensometer_data.m</code>, and use <code>POST_DIC_step3_impose_extracted_BC.m</code> to parse DIC results in a format suitable for imposing BC later in this script. Finally, comparison between experimental and computed displacements is provided by <code>POST_DIC_step4_modular_comparison_experiments.m</code>. (Note that the particular files need to be manually provided in the “Compute deformation process” part of <code>POST_DIC_step4_modular_comparison_experiments.m</code>.)</p> <p><strong>Experimental data</strong></p> <p>This folder contains data from (i) an unixaial tension test of a dogbone specimen (both from a MTS loading machine and DIC data) and (ii) two measurement sessions extracting the L-shape domain responses using DIC (<code>20_11_30 - Hluzek_Elka_newassemblyplan</code> and <code>21_04_12 - Hluzek_ Elka_quarters</code> with lower loading threshold). For post-processing, see the above-mentioned <code>POST_DIC_{...}.m</code> scripts. <code>*.mat</code> files present directly in <code>./experiments/</code> folder were obtained and are need by those scripts.</p> <p><strong>3D models</strong></p> <p>The folder contains geometrical models for individual parts needed for robot-assisted assembly of module molds for casting. This includes:</p> <ol> <li>a silo extension to store more tiles (file <code>silo_extension.stl</code>),</li> <li>formwork modules around the main structure for the purpose of casting silicone (file <code>tile_formwork.stl</code>),</li> <li>all types of tiles for the inside structure (file <code>tile_inside_types.stl</code>),</li> <li>a spacer shaped for YuMi base to ensure correct distance of the silo and build plate (file <code>yumi_base_1.stl</code>),</li> <li>a spacer shaped for YuMi base to ensure correct distance of the silo and build plate (file <code>yumi_base_2.stl</code>),</li> <li>a spacer shaped for YuMi base to ensure correct distance of the silo and build plate (file <code>yumi_base_3.stl</code>),</li> <li>connection for spacers (file <code>yumi_base_4.stl</code>),</li> <li>spacer holding a silo and the build plate (file <code>yumi_base_5.stl</code>),</li> <li>YuMi grippers with extensions to hold the tiles (file <code>grippers_extend.st</code>).</li> </ol> <p><strong>Robotics</strong></p> <p>The folder contains a single file with a script created in RobotStudio (RobotWare Version: 6.08.01.00, SmartGripper Version: 3.55.0000.00) to assemble the plan with YuMi IRB 14000-0.5/0.5 left hand.</p> <p><strong>Acknowledgement</strong></p> <p>The related research, experiments, and code development were supported by the <a href="https://gacr.cz/en/">Czech Science Foundation</a>, project No. 19-26143X.</p>
METRIC - Multi-Eye To Robot Indoor Calibration Dataset
<p>The METRIC dataset comprises more than 10,000 synthetic and real images of ChAruCo and checkerboard patterns. Each pattern is securely attached to the robot's end-effector, which is systematically moved in front of four cameras surrounding the manipulator. This movement allows for image acquisition from various viewpoints. The real images in the dataset encompass multiple sets of images captured by three distinct types of sensor networks: Microsoft Kinect V2, Intel RealSense Depth D455, and Intel RealSense Lidar L515. The purpose of including these images is to evaluate the advantages and disadvantages of each sensor network for calibration purposes. Additionally, to accurately assess the impact of the distance between the camera and robot on calibration, we obtained a comprehensive synthetic dataset. This dataset contains associated ground truth data and is divided into three different camera network setups, corresponding to three levels of calibration difficulty based on the cell size.</p>
Datasets and images of publication: Additive manufacturing for self-healing soft robots
<p>This entry contains the images and data used for the publication: Additive manufacturing for self-healing soft robots (DOI: 10.1089/soro.2019.0081). The datasets are named after the image they refer to and are available under the CC-BYSA 4.0 International license.</p>
Perception Sensor Dataset For Bioinspired Landing Trajectories Of An Ornithopter Robot
<p>The dataset contains the measurements captured by several onboard sensors during the landing maneuvers of an ornithopter robot. Each dataset contains a ROS bag file with the sensor measurements, a file with the bioinspired trajectory, a file with the events generated by the simulated event-based sensor, and a README file with the instructions to use the dataset.</p> <p>The bioinspired landing trajectories are computed using Tau Theory. Each landing trajectory test was performed in a simulated scenario. The object models of each scene can be found in the /model/meshes folder of each scene. There are two testing scenes: (i) a warehouse and (ii) a refinery. The file object_pose.csv includes the position and orientation of each object in the scene. The sensor measurements were saved in rosbag file that contains a topic for each sensor measurement. The dataset includes information from the following simulated sensors:</p> <ul> <li>Velodyne HDL-32E</li> <li>Sonar sensor with a range of 20 m</li> <li>IMU</li> <li>Frame based monocular camera</li> <li>Event camera</li> </ul>
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.