FULL TEXT http://homepages.inf.ed.ac.uk/s03496...r%20report.pdf

First Year PhD Report
Extracting Motion Primitives from Natural Handwriting Data
Ben H Williams

Institute of Adaptive and Neural Computation
School of Informatics
University of Edinburgh
5 Forrest Hill, Edinburgh UK
ben.williams@ed.ac.uk
August 10, 2005

Abstract. Movement selection and control is a very difficult inverse dynamics problem for robotic control that humans and animals accomplish easily. For the past 10 years there has been a popular theory that biological movement is made up of sub-routine type blocks, or ‘motor primitives’, with a central controller timing the activation of these blocks, creating synergies of muscle activation. Using machine learning techniques, this project addresses the possibility of extracting useful, and repeatable motor primitives from handwriting data.