Abstract:
Dexterous manipulation involves many skills. In this talk I will describe some pieces of work from my lab, and some thoughts about how manipulation skills might be organised. I’ll cover grasping, active touch, active vision, and prediction. Two algorithmic themes will be products and modules.
Bio:
Jeremy Wyatt is Professor of Robotics and Artificial Intelligence at the University of Birmingham. He has published more than ninety papers, and edited three books. He was originally a theologian, changed direction with an MSc in AI, and got his PhD in machine learning from the University of Edinburgh in 1992. He is primarily interested in robot task planning, and robot manipulation.