This four-year Project Grant from the National Science Foundation Division of Information and Intelligent Systems provides $599,998 to the University of Michigan for research titled "Overcoming Epistemic Uncertainty to Plan with Learned Dynamics Models for Robotic Manipulation."
Funded under the Computer and Information Science and Engineering program (CFDA 47.070), the award will support the development of techniques to help robots plan complex manipulation tasks despite uncertainty in how objects will interact dynamically. By advancing methods for robots to learn physical models of their environment, the research aims to enable more flexible and autonomous operation in unstructured or changing conditions. The University of Michigan will leverage its expertise in artificial intelligence, robotics, and dynamics modeling to execute the work through August 2025.
Generated 1/7/24, 12:06 AM