D7.2 - Data Management Plan

HARTU will adhere to FAIR principles:

• Making data findable:
  • All open dataset, publications, and open-source software produced by HARTU will be identifiable and locatable using a persistent Uniform Resource Identifier (URI) and/or Digital Object Identifier (DOI) to ensure the content is easily and uniquely citable.
  • Metada will be provided to allow discovery.
• Making data accessible
  • Data defined as public will be deposited in a trusted repository such as Zenodo. Additionally, all the links will be posted on the HARTU website and public code will be available on GitHub repository.
• Making data interoperable
  • Partners will observe OpenAIRE guidelines for online interoperability.
• Data re-usability
  • All datasets created by HARTU will provide Readme files that describe the organization and type of data, metadata, variable definitions, units of measurement etc. HARTU project will use and generate several datasets.

HARTU will make them available to stakeholders unless some restrictions apply for commercial or privacy reasons. The type of datasets include:

• Labelled 2D images and 3D point clouds of objects and environments, some of them generated with simulation tools and the rest by means of real data acquisition.
• Time-series data acquired from robots’ sensors. This includes trajectories of robot end effector pose, twist, spatial acceleration, and contact wrenches, as well as position, velocity, and acceleration of the robot joints.
• CAD of the workpieces and the positions of the corresponding contact points.
• Data on electro-adhesives performance (force vs displacement vs voltage) for different objects, data on surface properties of the objects and force-displacement data from the contact/force sensors.