Project

VeryHuman

Learning and Verifying Complex Behaviour of Humanoid Robots

Learning and Verifying Complex Behaviour of Humanoid Robots

  • Duration:
  • Application fields
    Other

The validation of systems based on deep learning for use in safety-critical applications proves to be inherently difficult, since their subsymbolic mode of operation does not provide adequate levels of abstraction for representation and proof of correctness. The VeryHuman project aims to synthesize such levels of abstraction by observing and analysing the behaviour of upright walking of a two-legged humanoid robot. The theory to be developed is the starting point for the definition of an appropriate reward function to optimally control the movements of the humanoid by means of enhanced learning, as well as for verifiable abstraction of the corresponding kinematic models, which can be used to validate the behaviour of the robot more easily.

Partners

Cyber Physical Systems (CPS), DFKI Robotics Innovation Center (RIC), DFKI

Sponsors

Federal Ministry of Education and Research (BMBF)

01IW20004

Federal Ministry of Education and Research (BMBF)

Images

[Translate to English:]

Publications about the project

Carlos Mastalli, Olivier Stasse, Heiner Peters, Vinzenz Bargsten, José de Gea Fernández, Frank Kirchner, Shivesh Kumar

In: 2020 IEEE-RAS 20th International Conference on Humanoid Robots (Humanoids). IEEE-RAS International Conference on Humanoid Robots (Humanoids-2020) July 19-21 Munich/Virtual Germany Pages 400-407 IEEE 7/2021.

To the publication
Andreas Mueller, Shivesh Kumar

In: 2021 IEEE International Conference on Robotics and Automation (ICRA). IEEE International Conference on Robotics and Automation (ICRA-2021) May 30-June 5 Xi'an China IEEE 6/2021.

To the publication

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz