Event Location: NSH 1507
Bio: Michael Zillich received his diploma in Mechatronics from Johannes Kepler University Linz in 1998 and his PhD in Electrical Engineering from Vienna University of Technology in 2007. Michael Zillich is (co)-author of over 80 publications and served on the program committees of a number of international conferences and as reviewer for international journals. His research interests are in computer vision (object segmentation, attention, recognition and tracking) with a focus on integrated cognitive systems and robotics. He is currently leading the European project SQUIRREL and nationally funded project InSitu and is PI in the European project STRANDS. He is also co-founder of Blue Danube Robotics, a company developing assistive robotic companions to support physically handicapped people in their daily lives.

Abstract: I will first give an overview of work done in our Vision for Robotics (V4R) lab over the past years, on object segmentation, recognition, classification, tracking, attention as well as grasping, and how that tied into various cognitive robotics projects. I will then address what I consider some open meta problems in robotics today: long term autonomy (moving from ran-once demos to months of autonomous operation), clutter (dealing with the spatial and temporal complexity in all aspects of real world scenarios), and – let’s face it – humans (offering a purposeful, convincing and safe user experience).