Motion planning for complex robotic systems is challenging due to the many degrees of freedom involved, as well as a variety of constraints induced by obstacles and robot kinematics. In this talk, I give an overview of the Covariant Hamiltonian Optimization and Motion Planning (CHOMP) algorithm, which can quickly generate smooth, high-quality paths for robots in tasks such as legged locomotion and mobile manipulation. Unlike some previous approaches which use randomized search followed by path smoothing, CHOMP is capable of directly solving for trajectories in a unified optimization framework. The talk will focus on two example systems: rough terrain locomotion for a quadruped robot, and highly constrained manipulation in a humanoid door opening task.
Bio: Matt Zucker is an Assistant Professor of Engineering at Swarthmore College, where he teaches courses in robotics, computer vision, and digital systems. In 2011, he received his Ph.D. from the Robotics Institute at Carnegie Mellon University, where he worked on motion planning for rough terrain locomotion. His research focuses on leveraging numerical optimization and machine learning techniques in order to improve planning speed and quality. Before graduate school, Matt worked from 2000-2005 writing software for autonomous underwater vehicles at Bluefin Robotics Corporation in Cambridge, MA.