The ability for a robot to plan its own motions is a critical component of intelligent behavior. The algorithmic questions that underpin motion planning have fascinated computer scientists for decades, due to inherent computational complexity and innumerable ways in which optimal decisions can be approximated. But recently, the application of robots into complex real-world scenarios, such as autonomous driving, are challenging the assumptions of classical theory. Where do models come from? What is the value of planning when plans invariably go wrong? Can optimality be adequately defined? Can we just throw machine learning at the problem and hope it disappears? This talk will outline a modern systems perspective of the role of planning, and how research in motion planning is evolving to suit this new role. Examples will be demonstrated on a diverse range of systems including warehouse automation, legged robot locomotion, and human-controlled robotic avatars.
Kris Hauser is an Associate Professor in the Department of Computer Science and the Department of Electrical and Computer Engineering at the University of Illinois at Urbana-Champaign. 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 then joined the faculty of Duke University from 2014-2019. Prof. Hauser is a recipient of a Stanford Graduate Fellowship, Siebel Scholar Fellowship, Best Paper Award at IEEE International Conference on Humanoid Robots 2015, the NSF CAREER award, and two Amazon Research Awards. He also works as a consultant for Google's autonomous driving company, Waymo.