Matt Zucker

Spring 2024 Office Hours: Tue 3:30-4:30 PM, Fri 10:30 AM-noon, and by appointment

Office: Singer 235 | mzucker1@swarthmore.edu

I am an associate professor in the Engineering department at Swarthmore College.

As an undergraduate, I majored in Cognitive Science at Vassar College, where I first began to appreciate taking a multidisciplinary approach to AI and robotics. Before grad school I worked for 5 years at Bluefin Robotics writing software for autonomous underwater vehicles, starting out as an undergraduate intern and departing as a senior software engineer. In 2005, I started my PhD at the Robotics Institute at Carnegie Mellon University, where I studied behavior generation for legged robots and mobile manipulators. My dissertation research focused on combining numerical optimization and machine learning to improve the capabilities of robot motion planning software.

I've been at Swarthmore since 2010. Although I still enjoy robotics research, I am also interested in a number of related areas such as computer graphics, laboratory automation for biology researchers, as well as rapid prototyping and CNC machining.

Here is my CV. You might also want to look at my coding blog.

Here is some advice for prospective graduate students, and related advice for students seeking recommendations from me.


Courses taught


Selected publications

Swarthmore student co-authors are highlighted. Please visit my Google Scholar page for a more complete listing.

cube-to-sphere example
Cube-to-sphere projections for procedural texturing and beyond
M. Zucker and Yosuke Higashi '18. Journal of Computer Graphics Techniques, 2018.
DRC-HUBO lifting heavy truss
Planning heavy lifts for humanoid robots
M.X. Grey, S. Joo, and M. Zucker. Proc. IEEE-RAS Int’l Conf. on Humanoid Robotics, 2014.
robot joint angle plots
Multigrid CHOMP with local smoothing
Keliang He '13, Elizabeth Martin '13, and Matt Zucker. Proc. IEEE-RAS Int’l Conf. on Humanoid Robotics, 2013.
CHOMP in action
CHOMP: Covariant Hamiltonian Optimization and Motion Planning
Matt Zucker, Nathan Ratliff, Anca D. Dragan, Mihail Pivtoraiko, Matthew Klingensmith, Christopher M. Dellin, J. Andrew Bagnell, and Siddhartha S. Srinivasa. International Journal of Robotics Research, May 2013.
learning preferences for footstep planning
Optimization and Learning for Rough-Terrain Legged Locomotion
Matt Zucker, Nathan Ratliff, Martin Stole, Joel Chestnutt, J. Andrew Bagnell, Christopher G. Atkeson, and James Kuffner. International Journal of Robotics Research, February 2011.
robot opening door
Continuous trajectory optimization for autonomous humanoid door opening
Matt Zucker, Youngbum Jun, Brittany Killen, Tae-Goo Kim, and Paul Oh. Proc. IEEE Int'l Conf. on Technologies for Practical Robot Applications (TePRA), 2013.
marble maze
Reinforcement Planning: RL for Optimal Planners
Matt Zucker and J. Andrew Bagnell. Proc. IEEE Int'l Conf. on Robotics and Automation, 2012.
trajectory optimization
CHOMP: Gradient Optimization Techniques for Efficient Motion Planning
Nathan Ratliff, Matt Zucker, J. Andrew Bagnell, and Siddhartha Srinivasa. Proc. IEEE Int'l Conf. on Robotics and Automation, May, 2009.


Features/Press appearances


Videos

Opening doors with the DRC-HUBO robot, using Multigrid CHOMP:

Getting Hubo walking during spring break 2013:

Robot solving a marble maze toy, June 2011

Rough terrain walking using the LittleDog quadruped robot: