Planning for Many Robots and Objects

With recent advances in Robotics and more broadly, Artificial Intelligence, it is foreseeable that robot will gradually step into every corner of our lives in the foreseeable future, addressing a variety of needs, e.g., helping with order fulfillment at large warehouses, assisting small businesses with product processing, or doing daily chores for the elderly/handicapped. Toward realizing these advances, however, robots must be able to perform planning and manipulation tasks efficiently and intelligently. In this talk, I will discuss some of our contributions in the domain of multi-robot coordination and robotic manipulation. On the topic of multi-robot coordination, I will highlight our recent theoretical/algorithmic breakthrough, Rubik Table algorithms, that enable a large number of robots to efficiently plan collision free paths in very dense settings. On the topic of robotic manipulation, we have examined a variety of long-horizon manipulation tasks including rearrangement, declutter, and object retrieval from clutter, for which we develop efficient algorithms, including both combinaltorial and data-driven ones, that compute natural and high-quality plans.

Post Talk Link:  Click Here

Passcode: @F7mX.v^