Abstract: A key component of intelligent behavior in robots is motion planning, which asks to generate continuous motions to achieve specified tasks while respecting constraints. This talk introduces recent achievements and research directions in optimal motion planning. It then proceeds beyond algorithms for solving single problem instances to seek a deeper understanding of the global structure of families of related problems. Armed with this global understanding, we can design algorithms that reason about changes to constraints, plan simultaneously across spaces of problem instances, or use adapt experience between problems. These approaches lead to faster and higher quality plans, which are in some cases fast enough to be suitable for inner loops of feedback controllers. Moreover, global structure can be embedded into machine learning models that learn optimal controllers with substantially higher accuracy than black box approaches (e.g., deep neural networks). This research has been inspired and informed by projects on a diverse range of robot systems, including mobile manipulators, humanoids, legged robots, warehouse automation, and autonomous vehicles.
Bio: Kris Hauser is an Associate Professor at the Pratt School of Engineering at Duke University with a joint appointment in the Electrical and Computer Engineering Department and the Mechanical Engineering and Materials Science Department. He received his PhD in Computer Science from Stanford University in 2008, bachelor's degrees in Computer Science and Mathematics from UC Berkeley in 2003, and worked as a postdoctoral fellow at UC Berkeley. He then joined the faculty at Indiana University from 2009-2014, where he started the Intelligent Motion Lab, and began his current position at Duke in 2014. He is a recipient of a Stanford Graduate Fellowship, Siebel Scholar Fellowship, Best Paper Award at IEEE Humanoids 2015, and an NSF CAREER award. His research interests include robot motion planning and control, semiautonomous robots, and integrating perception and planning, as well as applications to intelligent vehicles, robotic manipulation, robot-assisted medicine, and legged locomotion. http://people.duke.edu/~kh269/