Objective of workshop

Physical interactions with objects and the environment comprise a plethora of dexterous movements, ranging from tool making, playing a musical instrument, to surgical actions. Human skill displays not only low levels of variability while maximizing speed and accuracy, higher-dimensional actions also exploit the null space to be less sensitive to perturbations and noise. Unlike locomotory activity and grasping, such manual skills require extensive and continued practice to reach and maintain high levels. In robotics fine control of manual actions remains an open challenge. While robots can grasp a variety of objects, they remain fairly clumsy when it comes to manipulate the object once in hand. Robots are also ill at ease when it comes to manipulate objects with uneven inertia and friction, and objects that can deform. Learning has become an unavoidable step to achieve more manual dexterity in robotics. This workshop brings together researchers from human motor control and from robotics to answer the following questions: How do humans achieve such manual dexterity? What kind of practice schedules can shape these skills? Can some of these strategies be transferred to robots? To which extent is robot manual skill limited by the hardware, what can be learned and what cannot?

CFP Abstracts and demos

We call for 4-pages long abstracts on any of the following topics:
  • Algorithms for learning of manipulation skills in robots
  • Studies on control and learning of manual skills in humans
  • Implementation or validation of learning or adaptive controllers for robot manipulation
  • Biologically inspired controller for adaptive manipulation in robots
We also call for 1-page description of demos of:
  • learning algorithms in simulation or in real robots
  • methodology to monitor learning in humans
4-pages abstract and 1-page demo description must be sent by April 20 2020 to Dagmar Sternad and Aude Billard. Notification of acceptance will be sent out by April 22, 2020.

Program (tentative)

  • 8:45-9:00: Introduction by organizers
  • 9:00-10h30:
    Session 1: Intrinsic mechanics and finger coordination
    Madhusudhan Venkadesan: TBA
    Alberto Rodriguez: TBA
    15 minutes spotlights
    15 minutes panel discussion
  • 10h30-11h: Coffee break + Poster Session + Demos
  • 11h-12h30:
    Session 2: From grasping to manipulation
    Frederic Danion Adaptation of manual tracking to challenging cursor-target mappings
    Oliver Brock Low-dimensional representations of manipulation actions for learning
    15 minutes spotlights
    15 minutes panel discussion
  • 12h30-13h30: Lunch break
  • 13h30-15h00:
  • Session 3: Learning skills
    Jeannette Bohg Learning to scaffold the development of robotic manipulation skills
    Dagmar Sternad Learning and control in skilled interactions with objects: A task-dynamic approach
    15 minutes spotlights
    15 minutes panel discussion
  • 15h-15h30: Coffee break + Poster Session + Demos
  • 15h30-17h00:
    Session 4: Similarities and Differences in manipulation skills
    Andrea d'Avella Inter-individual differences in real-life motor skills and muscle synergy learning
    Antonio Bicchi From individualities in prosthetics to synergies for robotics hands and back again
    15 minutes spotlights
    15 minutes panel discussion
  • 17-18h00:
    Closing lecture: Etienne Burdet Augmented manipulation in humans and robots
    Closing discussion - organizers