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.
2,639
datasets available to search
ShareScore release 0.7.1
Dataset results
2,639 results for “robotic”
Raw data employed to perform the algorithm used in the scientific paper: "Kinematic reconstruction of the human arm joints in robot-aided therapies with Hermes robot"
<p>This file contains the raw data necessary to perform the algorithm introduced in the scientific paper:</p> <p>PAPER: Kinematic reconstruction of the human arm joints in robot-aided therapies with Hermes robot</p> <p>Authors: Arturo Bertomeu-Motos, Ricardo Morales, Luis D. Lledó, Jorge A. Díez, Jose M. Catalan, Nicolas Garcia-Aracil.</p> <p>Conference: EMBC 2015, IEEE 37th International Conference in Medicine and Biology Society, August 2015.</p> <p>Raw data acquired necessary to perform thee algorithm introduced in this paper.</p> <p>a) Robot Joints: Robot joints generated to develop the simulation, in radians (j1-j7 colums). This robot is referenced in the paper.<br> b) Direct Upper Limb Joints: Upper limb joints generated to develop the simulation, in radians (q1-q7 columns). This data is used to simulate the accelerometer value.</p>
Raw data corresponding to the scientific paper: "A modular telerehabilitation architecture for upper limb robotic therapy" (Advances in Mechanical Engineering 2017, Vol. 9(1) 1-13)
<p>Acquired raw data necessary to implement the adaptive control strategy grounded on multimodal information.<br> In addition, raw data for the computation of the communication parameters needed for the assessment of the implemented telerehabilitation architecture are provided.</p> <p>a) End-effector positions and velocities (x, y, vx, vy) in three conditions: healthy (Fig 9) and constraint simulated stroke behaviour (Fig 10) without robotic assistance and simulated stroke behavior with robotic assistance (Fig 11)</p> <p>b) Performance indicators and control parameters for all the recruited subjects in both conditions healthy behaviour and simulated stroke behaviour (Fig 12a and Fig 12b)</p> <p>c) Computational values for evaluating telerehabilitation performance (Table 1)</p> <p> </p> <p> </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>
IMU and marker-based optical motion capture from a humanoid robot
<p>The motion capture contains walking trials from the lower body of the humanoid robot Reem-C from Pal Robotics (Barcelona, Spain). Seven IMUs were attached on the foot, lower leg, upper leg and pelvis segments. IMU data was collected at 100 Hz. Moreover, the robot motion was captured with a marker-based optical system (Qualisys AB, Göteborg, Sweden) at 150 Hz. The focus of the dataset was mainly walking. There are three trials, each with a length of about 6.5 minutes.<br>The dataset contains the definition of the skeleton (segment lengths and coordinate locations), the actual IMU readings and the pose or kinematics from the optical system.</p>
Extending the executability of assembly task poses by robot through end-effectors
<p><strong>Extending the executability of assembly task poses by robots through end-effectors</strong></p> <p><br><em>Aline Kluge-Wilkes, Presley Demuner Reverdito</em></p> <p>The following data set was created during the validation of a proposed method to evaluate assembly station formations considering the executability of assembly tasks, including the effects of equipped end-effectors on robots.</p> <p>The underlying paper can be found at: https://doi.org/10.1007/978-3-031-34821-1_58 </p> <p>The underlying source code can be found at: https://git-ce.rwth-aachen.de/wzl-mq-public/iop/ws-b2.iv_formation-planning-of-mobile-robots/end-effector-dependent-executability</p> <p>In dependence on the geometries and degrees of freedom of the equipped end-effectors on the robot, a single task pose is transferred into an area of feasible task poses. Determining the executability of the task, the resulting representation of the feasible task poses is overlapped with the robot's workspace. If an overlap occurs, it can be assumed that there is a feasible robot configuration to execute the allocated assembly task. The proposed method is exemplified on a UR10 equipped with a screwdriver, and distributed task poses. The proposed method quantifies the end-effector's effect on the executability of assembly tasks and provides a means of determining the feasible base placement of robots in changeable assembly stations. Therefore, the method lays the foundation for automated formation planning in assembly stations.</p> <p><strong>Design of Experiments:</strong></p> <p>The published data here results from a conducted series of experiments structured as a full-factorial design of experiments. The following variables and expressions / data points of those variables were chosen:</p> <p>Tasks ("GoalPose") poses as [x, y, z, x-quaternion, y-quaternion, z-quaternion, w-quaternion]:</p> <ul> <li> Task 1 = (0.416m ,-0.399m, 0.764m, 0.071, 0.703, 0.134, 0.694)</li> <li> Task 2 = (0.438m, -0.647m, 0.816m, 0.005, 0.707, 0.068, 0.704)</li> <li> Task 3 = (0.448m, -0.755m, 0.945m, -0.243, 0.664 ,-0.183, 0.683)</li> </ul> <p>Dimension of the tools ("DimensionOfTool") as [x, y, z]: </p> <ul> <li> (0.1m, 0.2m, 0m)</li> <li> (0.2m, 0.1m, 0m)</li> </ul> <p>Robot model ("RobotModel"): </p> <ul> <li> UR10 (https://www.universal-robots.com/de/produkte/ur10-roboter/)</li> <li> UR5 (https://www.universal-robots.com/products/ur5-robot/)</li> </ul> <p>Resolution of reachability map ("ReachMap"): </p> <ul> <li> 0.05m</li> <li> 0.08m</li> <li> 0.1m</li> </ul> <p>IK solver ("IKSolver"): </p> <ul> <li>TRAC-IK <ul> <li>documented in: http://docs.ros.org/en/kinetic/api/moveit_tutorials/html/doc/trac_ik/trac_ik_tutorial.html </li> <li>source: https://bitbucket.org/traclabs/trac_ik/src/master/</li> </ul> </li> <li>KDL solver <ul> <li>https://docs.orocos.org/kdl/overview.html</li> </ul> </li> </ul> <p>Based on these definitions, a fully factorial design was implemented, resulting in 72 required tests (2³ × 3²). These tests were randomly organized into a single experimental block.</p> <p><strong>Test execution:</strong></p> <p>The function "[calcForAllTasks](https://git-ce.rwth-aachen.de/wzl-mq-public/iop/ws-b2.iv_formation-planning-of-mobile-robots/end-effector-dependent-executability/-/blob/main/code/main.py?ref_type=heads)" on the file main.py was created to generate 72 files, each containing 10 poses of the circle of possible poses for the robot with a calculated reachability map. These poses were manually tested using MoveIt! (https://moveit.ros.org/) and using two different IK solvers, KDL and TRAC-IK.</p> <p>There were now four packages: ur5_kdl, ur5_trac_ik, ur10_kdl, and ur10_trac_ik. Each package had its own group_name, which was the same as the package name. For each test, it was necessary to run the MoveIt! file of the robot with the correspondent IK solver, and the [move_group_python_interface.py](https://git-ce.rwth-aachen.de/wzl-mq-public/iop/ws-b2.iv_formation-planning-of-mobile-robots/end-effector-dependent-executability/-/blob/main/code/move_group_python_interface.py?ref_type=heads) with the desired pose. On the Python code, the goal pose was set, and using the MoveGroupPythonIntefaceTutorial class, Moveit! attempt to move the robot to the desired pose. If the pose is reachable, the reachability index was set to 1. Otherwise, it was set to 0, and the terminal output would show "ABORTED: No motion plan found. No execution attempted."</p> <p>After simulating all poses, the reachability average of each test was calculated, along with the sum of reachable results. The reachability index ranged from 0 to 100, and the reachable column ranged from 0 to 10. This result can be found in the file DoE-tests-and-results.xlsx.</p> <p> </p> <p>The file "DoE-tests-and-results.xlsx" provides an overview of all experiments in its first table. The creation order ("StdOrder") and the order in which the experiments were conducted are given ("RunOrder"). The following columns in the first table indicate the expressions of the variables as explained above (GoalPose, DimensionOfTool, RobotMode, ReachMap, IKSolver). Next, the results of the experiments are given: the average reachability index per tested goal pose and the indication of how many of the ten discrete robot flange poses per goal pose are reachable by the robot (Reachability Index, Reachable). </p> <p>The following 72 tables each provide the ten robot flange poses per goal pose per experiment as [x, y, z, x-quaternion, y-quaternion, z-quaternion, w-quaternion]. </p> <p>The 72 .csv files contain the same information as the 72 tables in the Excel file but contain the information as it was transferred during execution. </p> <p> </p> <p> </p> <p>After all simulations, it is possible to conclude that the reachability depends on the position of the task and the model of the robot. For example, UR5 could not reach task 3, while the UR10 had a high value of reachability for it due to the fact that the UR10 has a bigger workspace.</p> <p>-------------------------------------------------------------------------------------------------------------------------------------------------------</p> <p>Acknowledgement:</p> <p>Funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany's Excellence Strategy - EXC-2023 Internet of Production - 390621612.<br> </p>
Real-world grasp data of a dual-arm Yumi robot with a parallel gripper and suction cup end-effectors
<p>The attached txt file contains indexes to a cleaner subset of the data issued in the first version.</p> <p>Note: <br>Version 1 contains samples with failure cases due to environment constraints, which work well for the platform used in GraspAgent 1.0 (https://doi.org/10.1109/LRA.2024.3502066). However, this can degrade the performance if used on another platform with different constraints. To solve this, version 2 reports a subset of the raw data, excluding the failure modes due to environmental causes. </p>
Data for Publication: "Automated Investigation of Metal-Ligand Interactions by a Newly Established Robotic Workflow for Titrations"
<p>This dataset contains the whole primary and raw (original) data for the manuscript "Automated investigation of metal-ligand interactions by a newly established robotic workflow for titrations".</p>
High-throughput robotic titration using computer vision
<ul> <li> <p>An automated HTE robotic titration using a liquid-handling robot Opentrons(OT-2) and a standard webcam enables in-situ, affordable titration analyses.</p> </li> <li>Its modular design allows adaptability for materials chemsitry and integration into automated workflows, enhancing efficiency in chemical search.</li> </ul>
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>
Supramolecular Self-Healing Sensor Fiber Composites for Damage Detection in Piezoresistive Electronic Skin for Soft Robots
<p>Self-healing materials can prolong the lifetime of structures and products by enabling the repairing of damage. However, detecting the damage and the progress of the healing process remains an important issue. In this study, self-healing, piezoresistive strain sensor fibers (ShSFs) are used for detecting strain deformation and damage in a self-healing elastomeric matrix. The ShSFs were embedded in the self-healing matrix for the development of self-healing sensor fiber composites (ShSFC) with elongation at break values of up to 100%. A quadruple hydrogen-bonded supramolecular elastomer was used as a matrix material. The ShSFCs exhibited a reproducible and monotonic response. The ShSFCs were investigated for use as sensorized electronic skin on 3D-printed soft robotic modules, such as bending actuators. Depending on the bending actuator module, the electronic skin was loaded under either compression (pneumatic-based module) or tension (tendon-based module). In both configurations, the ShSFs could be successfully used as deformation sensors, and in addition, detect the presence of damage based on the sensor signal drift. The sensor under tension showed better recovery of the signal after healing, and smaller signal relaxation. Even with the complete severing of the fiber, the piezoresistive properties returned after the healing, but in that case, thermal heat treatment was required. With their resilient response and self-healing properties, the supramolecular fiber composites can be used for the next generation of soft robotic modules</p>
Fabrication of a Soft Robotic Gripper With Integrated Strain Sensing Elements Using Multi-Material Additive Manufacturing
<p>With the purpose of making soft robotic structures with embedded sensors, additive manufacturing techniques like fused deposition modeling (FDM) are popular. Thermoplastic polyurethane (TPU) filaments, with and without conductive fillers, are now commercially available. However, conventional FDM still has some limitations because of the marginal compatibility with soft materials. Material selection criteria for the available material options for FDM have not been established. In this study, an open-source soft robotic gripper design has been used to evaluate the FDM printing of TPU structures with integrated strain sensing elements in order to provide some guidelines for the material selection when an elastomer and a soft piezoresistive sensor are combined. Such soft grippers, with integrated strain sensing elements, were successfully printed using a multi-material FDM 3D printer. Characterization of the integrated piezoresistive sensor function, using dynamic tensile testing, revealed that the sensors exhibited good linearity up to 30% strain, which was sufficient for the deformation range of the selected gripper structure. Grippers produced using four different TPU materials were used to investigate the effect of the Shore hardness of the TPU on the piezoresistive sensor properties. The results indicated that the <em>in situ</em> printed strain sensing elements on the soft gripper were able to detect the deformation of the structure when the tentacles of the gripper were open or closed. The sensor signal could differentiate between the picking of small or big objects and when an obstacle prevented the tentacles from opening. Interestingly, the sensors embedded in the tentacles exhibited good reproducibility and linearity, and the sensitivity of the sensor response changed with the Shore hardness of the gripper. Correlation between TPU Shore hardness, used for the gripper body and sensitivity of the integrated <em>in situ</em> strain sensing elements, showed that material selection affects the sensor signal significantly.</p>
Piezoresistive sensor fiber composites based on silicone elastomers for the monitoring of the position of a robot arm
<p>Combining conductive fillers like carbon black with elastomers allows the development of soft elastomer strain sensors that can reach very large elongations, an important requirement for many robotic applications. However, when the conductive filler is introduced in the polymer, significant stiffening occurs, affecting the mechanical properties, e.g. Young’s Modulus, of the soft structure. In this attempt, single piezoresistive fiber composites were successfully fabricated, without drastically increasing the stiffness. Two silicone elastomers that are widely used in robotic applications were examined as matrix materials. Furthermore, modeling the stresses exerted on the fiber inside the composite was successfully used to predict the detachment of fiber inside the matrix, observed by visual inspection. For the PDMS based composite, pre-straining improved sensor properties, which could be confirmed for the monitoring of the movement of the crane robot. The results showed that the pre-strained piezoresistive sensor fiber-matrix composites positions of the robot crane can be monitored even at low strains.</p>
Multi-material 3D Printing of Thermoplastic Elastomers for Development of Soft Robotic Structures with Integrated Sensor Elements
<p>Embedded sensing can benefit soft robots with the ability to interact with their environment but producing embedded soft sensors can be challenging. Multi-material Fused Deposition Modeling (FDM) additive manufacturing allows producing complex structures, by combining more than one kind of polymeric material. For multi-material FDM, conductive thermoplastic elastomer filaments have been developed. This allows the printing of flexible functional structures, based on thermoplastic elastomer structures with conductive paths that are of great interest for stretchable electronics and soft robotic applications. In this study, stretchable piezoresistive elastomer strain sensor composites were successfully produced by using multi-material FDM. A piezoresistive thermoplastic elastomer was printed on the top of a nonconductive, flexible thermoplastic elastomer strip using FDM multi-material 3D printer. FDM elastomer filaments with different shore hardness as substrate materials for the gripper structure were used. The hardness of the elastomer affected the printability and the adhesion to the conductive elastomer material, which was used as a strain sensor material. The hardness affected the strain sensor properties too. The piezoresistive response, dynamic behavior, drift, relaxation and sensitivity of the printed multi-material strips were investigated by tensile tests. Soft robotic grippers with integrated sensing elements to detect deformation while touching the objective were selected as a case study. The soft grippers with the integrated sensors exhibited intelligent response by recognizing when they were griping a small or big object and when an obstacle was inhibiting their function.</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>
The Three Rs: Resolving Respiration Robotically in Shelf Seas
<p>Ocean gliders were deployed to conduct 'virtual mooring’ profiles at a study site in the seasonally stratified central Celtic Sea (station CCS, 49° 24’ N, 8° 36’ W) (see Fig. 1) during spring 2015 (6th April to- 28th April, decimal day 95 to 117) and summer 2015 (15<sup>th</sup> July and- 2nd August, decimal day 195 to 213). The integrated approach adopted in this study, combining ship based and glider measurements enabled estimates of spatial gradients while also minimising tidal aliasing that would likely be introduced by long spatial transects with the glider. A Slocum (Teledyne Webb Research, Falmouth, USA) Ocean Microstructure Glider (OMG, see Palmer et al., 2015 for full details) was equipped with a MicroRider microstructure package (Rockland Scientific International) to measure turbulentthe microstructure of velocity shear, a Seabird SBE42 CTD sensor to measure temperature, salinity and pressure, and an Aanderaa 4831 oxygen optode to measure O<sub>2</sub> (precision 0.2 µmol kg<sup>-1</sup>). Measurements were taken within 5 m of the bed and 2 m of the surface on most dives, with each yo-yo profile taking approximately 20 minutes. Glider salinity data was corrected for thermal inertia following Palmer et al. (2015). The glider AA4831 optode is known to experience severe lag across strong oxygen gradients, and therefore oxygen data was corrected where possible for optode membrane lag following Bittig et al. (2014). Where optode lag across the oxycline was too great and so not correctable using this method, it was omitted and oxygen data from coinciding CTDs was used. In comparison to other oxygen optodes, the AA4831 has been documented by various scientific studies as being an extremely stable optode with low detectable drift ( <0.5% yr<sup>-1</sup>) and high precision of <0.2 µmol kg<sup>-1</sup> (Kortzinger et al., 2004; Nicholson et al., 2008; Johnson et al., 2010; Champenois & Borges, 2012). Optode drift was calculated in this study by comparing discrete Winkler-analysed samples taken at deployment and recovery of the gliders, identifying a downward drift of 0.001% d<sup>-1</sup>, in close agreement with quoted manufacturer values.</p> <p>Glider sensors (temperature, salinity and ) were calibrated against nearby ship CTD profiles (CTD calibrated 1 month prior to cruise, SBE 43 precision = 2% of saturation) and discrete water samples collected within 3 hours and 2 km of glider deployment and recovery times and glider position, respectively, as part of the Shelf Sea Biogeochemistry programme (<em>RRS Discovery</em>, DY029 and DY033). Error estimates for the total change in (µmol kg<sup>-1</sup>) were calculated as the sum of the optode precision (0.2 µmol kg<sup>-1</sup>) and drift over the entire respective deployments (<0.1 µmol kg<sup>-1</sup>). Currents, tides, salinity and temperature were monitored throughout the glider deployments by a mooring at the CCS study site, which was equipped with an acoustic current profiler (ADCP), salinometer and thermistors that provided near-continuous data (Wihsgott et al., 2019; Ruiz-Castello et al., 2019).</p>
Supplementary materials (set 2 of 2) in support of "Signalling Emotions with a Breathing Soft Robot" (Data set and materials used for human-robot interaction experiment)
<p>Supplementary materials (set 2 of 2) in support of "Signalling Emotions with a Breathing Soft Robot" authored by Troels Aske Klausen, Ulrich Farhadi, Evgenios Vlachos, and Jonas Jørgensen.</p> <p>Contents of set 2:<br> - Data set and materials used for the human-robot interaction experiment and for data analysis</p> <p>Files:<br> - "Questionnaire.pdf": Questionnaire used for data collection.<br> - "Video links.txt": Weblinks to stimuli videos used.<br> - "Data set.xls": Collected raw data.<br> - "Matlab_DataAnalysis.mlx": Matlab script used to analyze raw data.<br> - "Linear_Arousal.png": Linear fit between the scoring of arousal and BPM.<br> - "Linear_Dominance.png": Linear fit between the scoring of dominance and BPM.<br> - "Linear_Pleasure.png": Linear fit between the scoring of pleasure and BPM.</p> <p>The experiment procedure is described in the paper.<br> The soft robot used for the experiment is open source and can be manufactured using design files available on Zenodo: 10.5281/zenodo.5565201</p>
The Robot Joint Torque Measurements for Accidental Collisions and Intentional Contacts
<p>This dataset contains the joint toque measurements of a robot manipulator (<a href="https://blog.robotiq.com/bid/64944/Collaborative-Robot-Series-KUKA-s-Light-Weight-Robot-4">KUKA LWR4+</a>) under accidental collisions and intentional contacts. It is specifically intended for the research study on robot collision detection, classification, diagnosis, or prediction. The dataset was recorded at <a href="https://www.ce.cit.tum.de/en/lsr/home/">Chair of Automatic Control Engineering</a>, <a href="https://www.tum.de/en/">Technical University of Munich</a>, Munich, Germany, by <a href="https://sites.google.com/view/zengjie-zhang/home">Dr. Zengjie Zhang</a>, under the supervision of <a href="https://www.ce.cit.tum.de/lsr/team/dozenten/dirk-wollherr/">Dr. Dirk Wollherr</a>, in 2017. Its detailed recording procedure is explained in the following work:</p> <p>[1] <strong>Zhang Z</strong>, Qian K, Schuller B W, and Wollherr D. An online robot collision detection and identification scheme by supervised learning and bayesian decision theory[J]. <em>IEEE Transactions on Automation Science and Engineering</em>, 2020, 18(3): 1144-1156.</p> <p>The dataset contains a number of external signal pieces of three classes: accidental collision (cls), with intentional manual contacts (ctc), and free from contacts (fre). Each signal piece lasts for 1.024s subject to the sampling rate 1kHz. Collisions or contacts occur at 0.256s of the signal pieces. The unit of the signal measurement is Nm. All the signals are recorded for the seven joints (#1 to #7) of the KUKA robot arm.</p> <p>The dataset is stored in .csv files. Each .csv file, containing the torque signal pieces for each class and each joint, is formed as an N by M matrix, where M = 1024 is the length of the signals and N is the number of signal pieces of the corresponding classes. For 'cls', N = 6960; for 'ctc', N = 7583; and for 'fre', N = 14098. Refer to the 'ReadMe.md' file for how to import the data to Python or MATLAB.</p> <p>This dataset is openly accessible for research work. Please cite this dataset and reference [1] if you publish the work based on them.</p>
Operator and Robot Pose Data
<p>This is a dataset of a .bag file format.</p> <p>The data included in this dataset are collected during a 4 hour shift of an operator. The operator is working alongside a collaborative cobot (COMAU AURA). During his shift, the operator is performing the assembly of a solar collector. The collector is attached on the robotic gripper and is manipulated by the cobot, while the operator proceeds with the assembly. For this certain bag file, the ROS message definitions are included:</p> <p>---</p> <p>ROS Message: geometry_msgs/TransformStamped</p> <p>This expresses a transform from coordinate frame header.frame_id to the coordinate frame child_frame_id</p> <p><strong>Header </strong>header</p> <p><strong>string </strong>child_frame_id # the frame id of the child frame</p> <p><strong>Transform </strong>transform</p> <p>---</p> <p>ROS Message: std_msgs/Header</p> <p>Standard metadata for higher-level stamped data types. This is generally used to communicate timestamped data in a particular coordinate frame.</p> <p>sequence ID: consecutively increasing ID</p> <p><strong>uint32 </strong>seq</p> <p>Two-integer timestamp that is expressed as:</p> <p>stamp.sec: seconds (stamp_secs) since epoch (in Python the variable is called 'secs')</p> <p>stamp.nsec: nanoseconds since stamp_secs (in Python the variable is called 'nsecs')</p> <p>time-handling sugar is provided by the client library</p> <p><strong>time </strong>stamp</p> <p>Frame this data is associated with</p> <p><strong>string </strong>frame_id</p> <p>---</p> <p>ROS Message: geometry_msgs/Transform</p> <p>This represents the transform between two coordinate frames in free space.</p> <p><strong>Vector3 </strong>translation</p> <p><strong>Quaternion </strong>rotation</p> <p>---</p> <p>ROS Message: geometry_msgs/Vector3</p> <p>This represents a vector in free space. It is only meant to represent a direction. Therefore, it does not make sense to apply a translation to it (e.g., when applying a generic rigid transformation to a Vector3, tf2 will only apply the rotation).</p> <p><strong>float64 </strong>x</p> <p><strong>float64 </strong>y</p> <p><strong>float64 </strong>z</p> <p>---</p> <p>ROS Message: geometry_msgs/Quaternion</p> <p>This represents an orientation in free space in quaternion form.</p> <p><strong>float64 </strong>x</p> <p><strong>float64 </strong>y</p> <p><strong>float64 </strong>z</p> <p><strong>float64 </strong>w</p>
High Payload Collaborative Robot - Joint States/Motor Current/TCP Force-Torque
<p>This dataset contains bag files, with data related to robot joint position, motor currents, robot tcp pose, robot tcp Force torque values etc. that were used for the design and development of a redundant collision detection for collisions with the robotic tool. There are also data with measurements from an external F/T sensor for the validation of the approach.</p>
Accelerometer and Force/Torque Sensor Measurements for Parameter and State Estimation of an Unknown Robot End Effector
<h1>Introduction</h1> <p>This dataset was created as part of a study on the development of an estimator for the contact wrench (force and torque) of an unknown robot end effector. A conference paper from this study has been submitted and accepted for the 2024 IEEE International Conference on Real-time Computing and Robotics (RCAR 2024) [1]. </p> <p>A force/torque sensor (FTS) was attached to the robot wrist, and the unknown end effector was attached to the FTS. An inertial measurement unit (IMU) was in turn attached to the end effector. The FTS measurement can be decomposed into the (1) sensor bias, (2) contact wrench, and the effects from (3) gravity, (4) inertia, (5) vibrations, and (6) noise. Estimation of the contact wrench requires that the remaining effects are compensated for. The FTS and IMU sensor biases, as well as mass and mass center of the unknown end effector, were estimated as described by Vougioukas [2]. His method requires FTS and IMU samples from 24 specific orientations of the sensors. See his paper for a description of this calibration method.</p> <p>The hardware used to generate this dataset were:</p> <ul> <li>KUKA LBR Med 14 serial robot (KUKA AG, Germany)</li> <li>ATI Gamma FTS (ATI Industrial Automation, Inc., USA)</li> <li>ATI Netbox (ATI Industrial Automation, Inc., USA)</li> <li>MPU6886 IMU (M5Stack, China) </li> <li>Arduino Mega 2580 with a W5500 Ethernet Shield </li> </ul> <h1>Method</h1> <p>The robot was used to move the end effector, FTS, and IMU such that a trajectory could be replicated with high precision and accuracy. The trajectory was a simple rotation about the FTS y-axis. This trajectory and the resulting measurements were performed three times. The sensor signals were sampled during each iteration when:</p> <ol> <li>The robot moved freely without any kind of disturbance (<strong>basline</strong>).</li> <li>The robot moved freely with gentle taps to the robot body, using a rubber hammer (<strong>vibrations</strong>).</li> <li>The robot moved with gentle taps to the body using the hammer, and with a manual force exerted on the end effector (<strong>vibrations and contact</strong>).</li> </ol> <p>The IMU signal was obtained by the Arduino Mega using I2C, and the signal was sent from the Arduino to the external PC using the ethernet shield. This setup resulted in <strong>a phase of the IMU signal by 8416 μs</strong>. This was compensated for in the offline analysis of the study on the contact wrench estimator [1]. The sensor samplig rates were different for each sensor; they were approximately 100 Hz for the robot controller (FTS orientation measurements), 700 Hz for the FTS, and 254 Hz for the IMU. The frequency for each signal can be obtained through the timestamps in the dataset.</p> <h1>Dataset</h1> <p>Each CSV file has a row which serves as the header, which labels the columns of each file. The following nomenclature of the column labels were used:</p> <p><strong>t </strong> - Timestep in microseconds. Epoch time. <br><strong>fx,</strong> <strong>fy, fz</strong> - The force components as measured by the FTS.<br><strong>tx, ty, tz </strong>- The torque components as measured by the FTS.<br><strong>ax, ay, az </strong> - The acceleration components measured by the IMU.<br><strong>gx,gy,gz </strong>- The direction of the gravitational vector in the FTS frame.<br><strong>r11, r12, r13, r21, r22, r23, r31, r32, r33 </strong>- The components of the rotation matrix that represents the FTS orientation in the world frame. (R_wf)</p> <p>The measurements from the FTS and IMU signals from the 24 orientations (as required for the calibration method described by Vougioukas [2]), are stored in <strong>0-calibration_fts-accel.csv</strong>. Additionally, the files <strong>0-steady-state_wrench.csv </strong>and <strong>0-steady-state_accel.csv</strong> contains the continuous sensor signal from the FTS and IMU, respectively, while they were at rest; these two files can be used to calculate the sensor signal variances.</p> <p>After calibration, each sensor signal was recorded independently and stored in a separate file from the other sensors. The raw (biased) values were stored. Each test iteration produced three files:</p> <ol> <li>The end effector/FTS/IMU orientation in <strong>[test_iteration]_orientation.csv</strong></li> <li>The unbiased wrench as measured by the FTS in [<strong>test iteration]_wrench.csv</strong></li> <li>The unbiased acceleration as measured by the IMU in <strong>[test_iteration]_accel.csv</strong></li> </ol> <p>The test iteration prefix for these files are: <strong>1-baseline</strong>, <strong>2-vibrations, </strong>and <strong>3-vibrations-contact, </strong>as described in the previous section "Method". To obtain the relative time between samples across the test iteration files ([]<strong>_orientation</strong>, []<strong>_wrench</strong>, and []<strong>_accel.csv</strong>), load each dataset and determine which has the earliest timestamped sample on the first row. Then, subtract this initial timestamp value from all timestamps across the files for the respective test iteration.</p> <p>Note that the IMU frame does not align with the FTS frame (<strong>_accel.csv</strong> vs <strong>_wrench.csv</strong>), the following table describes the rotation matrix R_fa which can be used to transform the acceleration measurements from the IMU frame {a} to the FTS frame {f}. </p> <p>R_fa = </p> <table> <tbody> <tr> <td>0</td> <td>-1</td> <td>0</td> </tr> <tr> <td>0</td> <td>0</td> <td>1</td> </tr> <tr> <td>-1</td> <td>0</td> <td>0</td> </tr> </tbody> </table> <h1>References</h1> <p>[1] A. Skrede, "A Linear Discrete Kalman Filter to Estimate the Contact Wrench of an Unknown Robot End Effector", Accepted for the 2024 IEEE International Conference on Real-time Computing and Robotics (RCAR), Ålesund, Norway, June 2024 </p> <p>[2] S. Vougioukas, “Bias Estimation and Gravity Compensation For Force-Torque Sensors,” in Recent Advances in Simulation, Computational Methods and Soft Computing. WSEAS Press, 2001, pp. 82–85. </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.