第46讲     第三十二期

报告时间：2023年4月20日（周四）09:30~11:00

腾讯会议：449-803-430  会议密码：687252

报告专家：Jingjin Yu 

专家单位：Rutgers University

报告题目：Optimal Long-Horizon Robotic Manipulation Planning

报告摘要：

With recent advances in Robotics and, more broadly, artificial intelligence and machine learning, it is foreseeable that robot helpers will gradually step into our everyday lives in the foreseeable future, addressing a variety of needs, e.g., assisting small businesses with product processing or doing daily chores for the elderly. To realize this, however, robots must be able to perform manipulation tasks efficiently and intelligently. In this talk, I will discuss some of our efforts in this direction, examining a variety of long-horizon manipulation tasks, including rearrangement, de-clutter, and object retrieval from clutter. We formally show that computing (time-)optimal solutions for these manipulation tasks are generally NP-hard. Nevertheless, we are able to develop efficient algorithms, including both combinatorial and data-driven ones, that compute natural and high-quality plans for these challenging manipulation tasks.

报告人简介：

Jingjin Yu is an Associate Professor in the Department of Computer Science at Rutgers University in New Brunswick. He received his B.S. from the University of Science and Technology of China and obtained his M.S. in Computer Science and Ph.D. in Electrical and Computer Engineering, both from the University of Illinois, where he briefly stayed as a postdoctoral researcher. Before joining Rutgers, he was a postdoctoral researcher at the Massachusetts Institute of Technology. He is broadly interested in the area of algorithmic robotics, focusing on issues related to optimality, complexity, and the design of efficient decision-making methods. He is a recipient of the NSF CAREER award and an Amazon Research Award.