Lockheed Martin Robotics Seminar Series
Grasping and Manipulation: Towards Humanlike Dexterity
Robotics Institute and
Computer Science Department
Carnegie Mellon University
Even after decades of research, robot dexterity still lags human dexterity by a great distance. Challenges include the high dimensionality of the space within which a robot with arms and hands can move, the complexity of the maneuvers that seem to be important in everyday life, and the need to cope with uncertainty. In this talk, I will present some reflections on how to deal with these challenges. The talk has three parts. In the first, I will look at the evolution of computer architecture over the past decades and give my thoughts on the implications of this evolution for planning robot actions. In the second, I will consider the question of how many different grasps a robot must master to be considered capable of human like performance. In the third part of the talk, I will explore grasping as a process with a variety of useful collision and contact events. Throughout the talk, I will present a mix of results from robot experiments and human subjects studies.
Nancy Pollard is an Associate Professor in the Robotics Institute and the Computer Science Department at Carnegie Mellon University. She received her PhD in Electrical Engineering and Computer Science from the MIT Artificial Intelligence Laboratory in 1994, where she performed research on grasp planning for articulated robot hands. Before joining CMU, Nancy was an Assistant Professor and part of the Computer Graphics Group at Brown University. She has received the NSF CAREER award for research on 'Quantifying Humanlike Enveloping Grasps' and the Okawa Research Grant for "Studies of Dexterity for Computer Graphics and Robotics."