Amy LaViers, Electrical and Computer Engineering graduate student, Georgia Tech
Magnus Egerstedt, Electrical and Computer Engineering professor, Georgia Tech
Jan. 27, 2012, 3:00 p.m.
Stauffer B, Room 125
How do you get a robot to do the disco? Or perform a cheerleading routine? These acts involve understanding two distinct movement styles. In this talk, we posit that “style of motion’’ can be taken into account in a principled manner: through discrete motion sequencing followed by trajectory generation for each motion. The talk explains how formal methods such as Linear Temporal Logic (LTL) and existing dance theory, namely that of Rudolf Laban, can be used to generate stylistic behavior. Distinct stylized motion sequences built from the same underlying building blocks animated on the NAO humanoid robot and in simulation will demonstrate the overarching objective of the research: to facilitate subtle degrees of control over systems through a useful parameterization for stylistic human movement.