Advances in Robot Task Planning and Grasping
In this talk I will sketch advances in the domains of mobile robot task planning, and robot manipulation. I'll give an overview of the field, and some examples of work from my own lab. This will include work on robot planning, common sense reasoning, and robot grasping. In addition I will try to thresh out which of these technologies are likely to be disruptive in the short to mid-term, and which will be gradually progressive, and which are still several technical steps away from success.
Jeremy Wyatt is Professor of Robotics and Artificial Intelligence at the University of Birmingham. He conducts research in autonomy, intelligent robotics, machine learning, robot manipulation, robot vision, robot task planning, robot motion planning, and decision making under uncertainty. He has worked at the University of Birmingham for 18 years, and previously obtained his PhD in Artificial Intelligence from the University of Edinburgh. Professor Wyatt has published more than 90 papers, edited three books, and his career research funding exceeds £10m. He has coordinated two major international research projects in robotics: CogX (35 researchers, 6 universities) on robot planning and learning in unstructured worlds, and PacMan (25 researchers) on robot manipulation. He has won three best paper awards, and gathered in excess of 1900 citations for his work. He works with a variety of industry partners on early stage technology transfer in logistics, robot manipulation, computer vision and autonomy. His work on robot manipulation has been featured in the media worldwide; he regularly gives public lectures, speaks at high profile events with business and government leaders, and is regularly interviewed in the national and international media on robotics and artificial intelligence. He has advised the UK and EU with regards to policy on robotics, and reviews for a variety of grant giving agencies around the world.