The ability for robots to do useful work in the real world has been a long standing problem with important applications in many areas including manufacturing, space exploration, and domestic assistance. One of the key challenges has been designing robots that can function robustly even when the state of the world is uncertain. In this talk, I will frame this as a planning under uncertainty problem where a robot must perform a task robustly while simultaneously gathering perceptual information as necessary. Planning under uncertainty is a general framework in which the objective is to move from an initial “belief state” where the system is very uncertain about the state of the world to one where the system is certain that it has achieved the task objectives. One of the hallmarks of planning under uncertainty is the ability to generate information-gathering strategies automatically as needed to solve a problem. I will propose a new approach to planning under uncertainty that is particularly well suited to a large class of difficult robotics problems defined over continuous spaces and over long time horizons. This algorithm is more computationally efficient than other approaches and has provable convergence and correctness guarantees. I will demonstrate the approach in the context of a difficult robot manipulation problem where the robot must simultaneously localize and grasp an occluded object whose position is initially unknown.
Robert Platt is an Assistant Professor in the Computer Science and Engineering Department at SUNY Buffalo. Prior to that, he was a research scientist at MIT in the Computer Science and Artificial Intelligence Laboratory. He holds a Ph.D. from the University of Massachusetts, Amherst.