This $272,290 three-year Project Grant from the National Science Foundation's Division of Computer and Network Systems will fund the development of visual tactile neural fields (VTNF) for active digital twin generation by robots. The grant supports research by The Trustees of the University of Pennsylvania under the NSF's $47.07 million Computer and Information Science and Engineering program.
The researchers will create a new VTNF data representation that allows robots to combine visual and tactile sensor data into a unified model of objects. This will enable robots to more effectively and efficiently generate accurate models of their environment and physical interactions. Key activities include developing the mathematical and computational tools to generate VTNF models, quantifying model uncertainty, and techniques for robots to learn search policies to create models quickly. The VTNF is intended to provide a flexible foundation for other researchers to enable faster robotic learning through more detailed understanding of surroundings and interactions.
Generated 1/6/24, 4:57 PM