PhD Thesis

Bimanual Robot Skills: MP Encoding, Dimensionality Reduction and Reinforcement Learning

Work default illustration


  • Started: 01/12/2011
  • Finished: 14/07/2017


This thesis discusses and tackle all the problems associated to the learning of
motor tasks by complex robots - e.g., bimanual robots- in an often unstructured or unmodelable
environment, as it can be the task of folding clothes. Due to limited amount of data samples in
robotic learning scenarios, such tasks require dimensionality reduction techniques so that such
complex problems can be solved.

The work is under the scope of the following projects:

  • SGR ROBÒTICA: Grup de recerca consolidat - Grup de Robòtica (web)
  • GARNICS: Gardening with a cognitive system (web)
  • IntellAct: Intelligent observation and execution of Actions and manipulations (web)