Building Generalizable Mobile Manipulation System

Abstract:

What does it take to build mobile manipulation systems that can competently operate on previously unseen objects in previously unseen environments? In this talk, I will present recent case studies on building two such systems: one for interacting with articulated objects using a wheeled manipulator, and another for whole-body object grasping using a humanoid robot. I will discuss findings from large-scale real-world evaluations and distill key principles for building generalizable mobile manipulation systems, including a central one: that the best results come from marrying classical robotics principles with large-scale learning, rather than treating them as competing paradigms, at least for now.

Speaker Bio:

Saurabh Gupta is an Associate Professor in the Department of Electrical and Computer Engineering, at the University of Illinois at Urbana-Champaign (UIUC). Before starting at UIUC in 2019, he received his Ph.D. from UC Berkeley in 2018 and spent the following year as a Research Scientist at Facebook AI Research in Pittsburgh. His research interests span computer vision, robotics, and machine learning, with a focus on building agents that can intelligently interact with the physical world around them.