Title:
Spatiotemporal Correspondence as a Metric for Human-like Robot Motion
Spatiotemporal Correspondence as a Metric for Human-like Robot Motion
Author(s)
Gielniak, Michael J.
Thomaz, Andrea L.
Thomaz, Andrea L.
Advisor(s)
Editor(s)
Collections
Supplementary to
Permanent Link
Abstract
Coupled degrees-of-freedom exhibit correspondence, in that
their trajectories in
uence each other. In this paper we
add evidence to the hypothesis that spatiotemporal corre-
spondence (STC) of distributed actuators is a component of
human-like motion. We demonstrate a method for making
robot motion more human-like, by optimizing with respect
to a nonlinear STC metric. Quantitative evaluation of STC
between coordinated robot motion, human motion capture
data, and retargeted human motion capture data projected
onto an anthropomorphic robot suggests that coordinating
robot motion with respect to the STC metric makes the
motion more human-like. A user study based on mimick-
ing shows that STC-optimized motion is (1) more often rec-
ognized as a common human motion, (2) more accurately
identi ed as the originally intended motion, and (3) mim-
icked more accurately than a non-optimized version. We
conclude that coordinating robot motion with respect to the
STC metric makes the motion more human-like. Finally, we
present and discuss data on potential reasons why coordi-
nating motion increases recognition and ability to mimic.
Sponsor
Date Issued
2011
Extent
Resource Type
Text
Resource Subtype
Pre-print
Proceedings
Proceedings