Robot learns to flip pancakes


The video shows a Barrett WAM 7 DOFs manipulator learning to flip pancakes by reinforcement learning. The motion is encoded in a mixture of basis force fields through an extension of Dynamic Movement Primitives (DMP) that represents the synergies across the different variables through stiffness matrices. An Inverse Dynamics controller with variable stiffness is used for reproduction. The skill is first demonstrated via kinesthetic teaching, and then refined by Policy learning by Weighting Exploration with the Returns (PoWER) algorithm. After 50 trials, the robot learns that the first part of the task requires a stiff behavior to throw the pancake in the air, while the second part requires the hand to be compliant in order to catch the pancake without having it bounced off the pan. Video credits: ————————– Dr. Petar Kormushev kormushev.com Dr. Sylvain Calinon http Affiliation: ———————- Advanced Robotics dept. Italian Institute of Technology (IIT) Link to publication: programming-by-demonstration.org Link to publication on PoWER algorithm by Jens Kober and Jan Peters: books.nips.cc



Note: This is a user submitted video under “User Submitted” Category. Although the submitted content is moderated, RealityPod does not guarantee the accuracy of information provided in posts of this particular category.

Comments
  1. r1ck13579 Reply
  2. GrimNeko Reply
  3. planmix Reply
  4. wylmau Reply
  5. Notouchtail Reply
  6. TheShiftingStudios Reply
  7. paperman972 Reply
  8. XxShOoTnDiExX Reply
  9. Shalek Reply
  10. ral008 Reply
  11. aby0ni Reply
  12. NothingToCompareItTo Reply
  13. wylmau Reply
  14. ikilleduku Reply
  15. pietzeekoe Reply
  16. Quaft1 Reply
  17. Quaft1 Reply
  18. Mikeadrenalinzone Reply
  19. electroum Reply
  20. mpikybroll Reply
  21. magnvss Reply
  22. WhimsicalUnited Reply
  23. WhimsicalUnited Reply
  24. redsoxxx1234 Reply
  25. Dirtfire Reply

Leave a Reply

Your email address will not be published. Required fields are marked *