Seminar: Autonomous Action and Interaction Under Uncertainty - Oct. 9
Zachary Sunberg
Assistant Professor, Smead Aerospace
Friday, Oct. 9 | 10:40 A.M. | AERO 111
Abstract: Autonomous aerospace vehicles can tackle challenges ranging from collecting scientific data in hazardous environments to defending against swarms of attacking drones in ways that are impossible for humans alone. Tasks like these require overcoming uncertainty in the environment and interacting strategically with other humans and artificially-intelligent systems. The partially observable Markov decision process (POMDP) and partially observable stochastic game (POSG) are mathematical models that can represent many types of uncertainty and strategic interaction. However, finding solutions for these problems is notoriously difficult. The Autonomous Decision and Control Lab (ADCL) has studied these problems from a variety of perspectives ranging from mathematical theory to tests with real humans and developed algorithms that can solve them in surprisingly efficient ways. This talk will describe the challenges that make this task difficult, some of our key discoveries, and some of the questions that we want to answer in the future.
Bio: Zachary Sunberg is an Assistant Professor in the Ann and H.J. Smead Aerospace Engineering Sciences Department and director of the Autonomous Decision and Control Lab. He earned Bachelors and Masters degrees in Aerospace Engineering from Texas A&M University and a PhD in Aeronautics and Astronautics at Stanford University. Before joining the University of Colorado faculty, he served as a postdoctoral scholar at the University of California, Berkeley. In 2024, he received the NSF CAREER award. His research is focused on safe and efficient operation of autonomous vehicles and systems on the ground, in the air, and in space. A particular emphasis is on handling uncertainty using the partially observable Markov decision process and game formalisms.