OGST
Alex Litoiu – 3rd year
Advisor: Brian Scassellati
Title: Personalized Instruction of Physical Skills with a Social Robot
Abstract: Socially assistive robots have taken an important role in tutoring children in symbolic domains, but using robots to teach or refine motor skills to children is underserved. Helping children to be more proficient in physical skills has been shown to be highly conducive to increased physical activity and improved health throughout adolescence and beyond.
Robots are uniquely well situated to observe physical movements, identify problems, prioritize which problems to address first, and to patiently communicate personalized advice to the student. We propose an architecture to coach physical skills, and focus on the second and third of these challenges - identifying problems with the movements, and prioritizing which to address first - as applied to the domain of shooting a basketball. We discuss the sensing platform that we have constructed, an algorithm to detect problems in the student’s motion, as well as a module that allows a robot to deliver advice through physical demonstrations.
We identify future research questions that remain unanswered, and discuss some preliminary approaches to studying these research questions.