RBE Professor Sonia Chernova Receives Young Investigator Award from the Office of Naval Research
This award is for untenured faculty who show exceptional promise for doing creative research.
Robotics Professor and Computer Science Assistnat Professor Sonia Chernova recently won the prestigious Young Investigator Award from the Office of Naval Research (ONR). The award is for untenured faculty "who show exceptional promise for doing creative research." This is the second significant early faculty honor for Sonia, who in 2012 won a five-year National Science Foundation CAREER Award, the most prestigious NSF award for young faculty. In her three-year ONR project, titled "Multi-Representational Learning from Demonstration through Sequential User Study Development," Sonia will extend her groundbreaking work aimed at developing methods by which non-experts can teach high-level tasks to robots without the need for programming. As robots are increasingly uses in military applications, it will be imperative that personnel without special skills be able to teach robots new skills on the fly in response to changing field conditions. Using long-term user studies conducted over the web, Sonia and her team will seek to develop a novel learning framework that combines supervised learning and reinforcement learning to enable robots programmed with only a library of primitive skills to learn from a combination of user demonstrations, corrections, guidance, and rewards.
April 30, 2014