Kyle H Wray

Biography

Kyle Hollins Wray leads the creation of autonomous robots that can plan, make decisions, navigate, and coordinate in the real world. He also researches theoretical models and practical algorithms for decision-making under uncertainty in real robots. Previously, Kyle was a Director of Engineering at Robust AI where he lead the fleet intelligence team, program managed six teams, designed product workflows, and presented to venture capitalists to successfully raise a $20 million Series A-1 funding round. He was also a Principal Researcher at the Nissan Advanced Technology Center - Silicon Valley where he led the creation of autonomous vehicle decision-making systems and hybrid electric vehicle energy management systems, managed multi-million-dollar projects, and presented to stakeholders including the CEO of Nissan. He was also a Visiting Scholar at Stanford University working with Mykel Kochenderfer where he co-authored a textbook entitled Algorithms for Decision Making that was published by MIT Press.

He earned his doctorate in computer science from the University of Massachusetts Amherst, advised by Shlomo Zilberstein, specializing in artificial intelligence and autonomous robots. Kyle earned his bachelor's and master's degrees in computer science, engineering, and mathematics from the University of Massachusetts Amherst and The Pennsylvania State University.

He is the recipient of the ICAPS Best Dissertation Award (2020), the runner-up Victor Lesser Distinguished Dissertation Award (2020), the Nissan Patent Research Award (2020 and 2018), and the Outstanding Graduate Research Award (2016).

He serves as an Associate Editor for IEEE Robotics and Automation Letters (RA-L) and an Area Chair for AAAI Conference on Artificial Intelligence (AAAI).

Please see his curriculum vitae (CV) for more details.

Academic Research Interests

Kyle’s research interests include decision-making, modeling, planning, and learning in the fields of both artificial intelligence and robotics. His work spans the design and analysis of theoretical models, the creation of efficient algorithms, and the practical applications of these decision-making approaches.

In artificial intelligence, he is particularly interested in the theoretical foundations of decision-making models based on MDPs, SSPs, POMDPs, and Dec-POMDP. In particular, he designs novel multi-objective and hierarchical models that unify prior models, and proposes scalable algorithms that solve them.

In robotics, he is particularly interested the goal of true long-term autonomy. In particular, he designs novel MDPs, SSPs, POMDPs, and Dec-POMDPs, with efficient algorithms to solve them online, that enable scalability, are robust, and provide safety guarantees.

His models and algorithms are deployed on Robust AI’s collaborative mobile robots and Nissan’s autonomous vehicles.

News