Matt Zucker
I am interested in developing planning and control algorithms for complex legged robots and mobile manipulators. I believe that high-level planning, which reasons over sequences of discrete behavior primitives, is the best way to plan for such systems. My work focuses on leveraging optimization and machine learning techniques, as well as re-using previous computation, in order to produce fast planning software.
I am a member of the Planning and Autonomy Lab, and I particpate in the DARPA Learning Locomotion Project.
This page has a list of publications as well as some recent research videos.
Publications
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.
Adaptive Workspace Biasing for Sampling Based Planners
Matt Zucker, James Kuffner, and J. Andrew Bagnell
Proc. IEEE Int'l Conf. on Robotics and Automation, May,
2008.
Multipartite RRTs for Rapid Replanning in Dynamic Environments
Matt Zucker, James Kuffner, and Michael Branicky
Proc. IEEE Int. Conf. Robotics and Automation, April, 2007.
Improved Motion Planning Speed and Safety using Regions of Inevitable Collision
Nicholas Chan, James Kuffner, and Matt Zucker
17th CISM-IFToMM Symposium on Robot Design, Dynamics, and
Control (RoManSy'08), July, 2008.
Approximating State-Space Obstacles for Non-Holonomic Motion Planning
Matt Zucker
tech. report CMU-RI-TR-06-27, May 2006.