Foundations of Robotics
Seminar, November 12, 2008
Time
and Place | Seminar Abstract
From Vibratory Manipulation to Self-Organization
Kevin Lynch
Department of Mechanical Engineering
Northwestern University
NSH 1507
Talk 4:00 pm
Traditional robot assembly is a sequential centralized top-down
process, where one or more robot manipulators place parts into a
growing assembly. Self-assembly, on the other hand, is a parallel
decentralized bottom-up process, where parts are designed to bond with
each other so that the desired assembly is an attractor. In both
cases, the "assembly program" is encoded in the control algorithms and
physical properties of the manipulators, environment, and the parts
themselves.
Between these centralized and decentralized extremes lies most current
and future industrial assembly. In this talk I will describe our work
at two points on this spectrum. In the first part of the talk, I will
describe the PPOD, the Programmable Parts-Feeding Oscillatory Device.
The PPOD uses six-degree-of-freedom vibration of a rigid plate to
create frictional force fields that can control the motion of several
parts simultaneously. In the second part of the talk, the parts are
"smart" and capable of communication, actuation, and locomotion. I
will describe a systematic design procedure to compile a desired group
behavior into decentralized communication and control laws for the
individual agents.
This work is joint with my students Tom Vose and Peng Yang and my
colleagues Randy Freeman and Paul Umbanhowar.
Kevin Lynch is Associate Chair and McCormick Professor of Teaching
Excellence in the Department of Mechanical Engineering at Northwestern
University. His research interests include robotic manipulation,
self-organizing systems, human-robot interaction, and bio-inspired
sensing and locomotion. He is a Senior Editor of the IEEE
Transactions on Robotics and a recipient of the NSF CAREER award, the
2001 IEEE Early Academic Career Award in Robotics and Automation, and
the 2007 SAE Teetor Educational Award.
The Robotics Institute is part of the School of Computer Science, Carnegie Mellon University.