School of
Information Technology and Electrical Engineering

Robot Motion Planning in Unstructured, Dynamic, and Human Environments

Prof Kris Hauser - Indiana University, BloomingtonFri, 27/07/2012 - 14:00
Dr Surya Singh

The past decade has seen major advances in algorithmic tools for the motion planning problem, which is the problem of coordinating many joints to achieve complex goals.  While computationally and theoretically sound, these tools are often restricted in their applicability to real-world robots - primarily, they assume precise system models and large computational budgets.  The objective of my research is to create tools that still operate on firm theoretical foundations, but with relaxed assumptions in order to handle real world complexity, unknowns, and uncertainty.  My graduate and postdoctoral work aimed to scale up planning to high-dimensional systems (up to 42DOF) in unstructured and unpredictable environments, and was applied to legged locomotion in highly uneven, steep terrain, object manipulation for humanoid robots, and robot-assisted surgery in deformable tissue.  While at Indiana University I turned my attention to safe, real-time integration of robots into human-centred environments.  Toward this end, I have worked on cooperative motion planning algorithms for crash prevention of human-controlled robots applied to two domains: teleoperation of industrial robots, and autonomous emergency maneuvering for passenger vehicles.  Videos of these results will be shown on real and simulated robots, ranging from the NASA ATHLETE six-legged lunar robot, the Honda ASIMO humanoid, steerable surgical needles, industrial robots and intelligent vehicles.


Kris Hauser 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's Automation Lab.  He has held his current position as Assistant Professor of Computer Science at Indiana University since 2009, where he directs the Intelligent Motion Lab.  Research interests include algorithms for robot motion planning and control, integration of planning and perception, and semiautonomous robots.  Applications of his research have included vehicle collision avoidance, robotic manipulation, robot-assisted medicine and legged locomotion in rough terrain.

Seminar Type: 

ITEE Research Seminar