I am interested in sensing strategies that acquire object shape and
configuration concurrently during manipulation. My broader interests
include the mechanics of manipulation, nonprehensile manipulation, parts
assembly, cooperating robots, planning under uncertainty, probabilistic
strategies, sensing strategies, and automatic planning.
Robotics Motivations
I am interested in making robots act purposefully and successfully in a
world in which most everything is uncertain. Sensors are noisy, actions
are imprecise, and objects are often in the wrong location. Despite
such obstacles to purposeful action, there are many tasks that can be
accomplished successfully. Humans, animals, and some machines are
proof. Providing robots with the ability to operate autonomously and
purposefully requires an understanding of how different tasks may be
accomplished by different repertoires of actions. Grasping, hitting,
and dropping are some actions that are useful in a robot's repertoire.
More exotic actions include shaking, twirling, and other actions that
randomize an object's state.
My work is motivated by several desires. First, I would like to program
robots more easily than is currently possible. Second, I would like to
understand the scope and limitations of autonomous systems, whether biological
or artificial. Third, I would like to reduce the complexity of design and
planning by codifying the design parameters required to achieve a given level
of automation. An underlying goal of my research is to understand the
relationship between sensing, action, and prediction. In the past, I have
explored various extreme points in this space. With Matt Mason I explored
sensorless strategies, for my thesis work I looked at randomized strategies,
and most recently I investigated fast-action minimal-sensing strategies. My
research draws on tools from geometry, mechanics, planning, and
stochastic processes.
I have collaborated with Dr. Gordon Rule in the Department of
Biological Sciences on a method for determining protein
structure homology from sparse NMR data. More broadly, I am
interested in geometric algorithms for understanding the structure and
dynamics of allosteric proteins. Of particular interest to me is the
extent to which topological shapes act as fingerprint identifiers of
proteins. One novel result of this work is a method for representing
and comparing proteins using line weavings.
I am interested in the interplay of action and sensing. This
research forms a component of the SToMP!
project, led by Professor
Robert Ghrist at UIUC, involving mathematicians and roboticists
from several institutions around the world.
Older Projects Still of Interest:
I am interested in understanding the interplay of manipulation and
locomotion.
Matt Mason,
Siddhartha Srinivasa,
and I investigated this problem using a mobile manipulator
called the
Mobipulator.
I am interested in protein homology, in particular determining
structural homology from sparse NMR data and modeling protein
structures in terms of line weavings. I worked together with Gordon
Rule in the Department of Biological Sciences.
I am interested in sensing strategies that acquire object shape
and configuration concurrently during manipulation. This line of
research began with the Ph.D. thesis of my student Yan-Bin Jia, extended into
my own thinking, and continued with my student Mark Moll.
Former Students
Siddhartha Srinivasa
(co-advised with Matt
Mason), Robotics.
Ph.D., August 2005. Thesis title: Control Synthesis for Dynamic
Contact Manipulation.
First post-Ph.D. job: Intel Research Laboratory, Pittsburgh.
Mark Moll, Computer
Science. Ph.D., July 2002. Thesis title: Shape Reconstruction
Using Active Tactile Sensors.
First post-Ph.D. job: Research Associate, Physical and Biological
Computing Group, Rice University, then at USC-ISI, now a Research
Scientist at Rice.
Jovan Popovic,
Computer Science. Ph.D., July 2001. Thesis title: Interactive
Design of Rigid-Body Simulations for Computer Animation.
First post-Ph.D. job: Assistant Professor, MIT.
Yan-Bin Jia, Robotics.
Ph.D., November 1997. Thesis title: Geometric and Dynamic Sensing:
Observation of Pose and Motion through Contact.
First post-Ph.D. post-CMU job: University of Minnesota. Now a
tenured professor at Iowa State.
Nina Zumel,
Robotics. Ph.D., January 1997. Thesis title: A Nonprehensile
Method for Reliable Parts Orienting.
First post-Ph.D. job: AI group, Stanford Research Institute.
She now has her own company, Quimba Software.
Tamara Abell, Robotics. M.S., December 1995, currently at Apple.
Thesis Committee Member
Joel Chestnutt, Ph.D., Computer Science, CMU, 2007.
Jason O'Kane, Ph.D., Computer Science, UIUC, 2007.
Guillermo Bermejo, Ph.D., Chemistry, CMU, 2007.
Keith Kotay, Ph.D., Computer Science, Dartmouth, 2003.
Alexander Grishaev, Ph.D., Chemistry, CMU, 2001.
Dongmei Zhang, Ph.D., Robotics, CMU, 1999.
Tom Ault, Robotics, CMU, proposed 1998.
Barry Brumitt, Ph.D., Robotics, CMU, 1997.
George Paul, Ph.D., Robotics, CMU, 1997.
J. Dan Morrow, Ph.D., Robotics, CMU, 1997.
Cheng-Hua Wang, Ph.D., Robotics, CMU, 1997.
David Simon, Ph.D., Robotics, CMU, 1996.
Srinivas Akella, Ph.D., Robotics, CMU, 1996.
Kevin Lynch, Ph.D., Robotics, CMU, 1996.
David Wettergreen, Ph.D., Robotics, CMU, 1995.
Sanjiv Singh, Ph.D., Robotics, CMU, 1995.
Sing Bing Kang, Ph.D., Robotics, CMU, 1994.
Prasad Chalasani, Ph.D., Computer Science, 1994.
Rudi Stouffs, Ph.D., Architecture, CMU, 1994.
Alan Christiansen, Ph.D., Computer Science, CMU, 1992.
Harry Kim, Ph.D., Robotics, CMU, 1991.
Abstracts of Selected Papers
(Click here
for a publication list auto-generated by the Robotics Institute.)
Siddhartha S. Srinivasa, Christopher R. Baker, Elisha Sacks,
Grigoriy B. Reshko, Matthew T. Mason, and Michael A. Erdmann.
Experiments with Nonholonomic Manipulation.Proceedings of the 2002 IEEE International Conference on Robotics
and Automation, Washington, DC.
Full paper (1.2MB, 6 pages).
Mark Moll and Michael Erdmann.
Uncertainty Reduction Using Dynamics. Proceedings of the 2000 IEEE International Conference on Robotics and
Automation, San Francisco, California.
Yan-Bin Jia and Michael Erdmann.
Pose and Motion from Contact. International Journal of Robotics Research, Vol. 18, No. 5, 1999.
Yan-Bin Jia and Michael Erdmann.
Pose from Pushing. Proceedings of the 1996 IEEE International Conference on Robotics
and Automation, Minneapolis, Minnesota, pp. 165-171.
We gratefully acknowledge support by the National Science
Foundation for this research.
Relevant support includes a Research Initiation Award IRI-9010686
and REU supplement, a Presidential Young Investigator Award
IRI-9157643, grant IRI-9213993 (with REU supplement IRI-9443084),
grant IRI-9503648 (with REU supplements IRI-9642850 and IRI-9741440,
and a Creativity Extension), grant IIS-9820180, and grant IIS-0222875.
Any opinions, findings, and conclusions or recommendations
expressed in this research are those of the author(s) and do not
necessarily reflect the views of the National Science Foundation.