Andrew Stout

Papers

Total Citations

H-Index

About

Andrew Stout’s research lies at the intersection of developmental robotics, intrinsically motivated learning, and human-robot interaction. His most influential work, “Intrinsically Motivated Reinforcement Learning: A Promising Framework for Developmental Robot Learning” (2005, 64 citations), helped establish a foundational paradigm for enabling robots to learn complex, open-ended behaviors without external rewards. This framework, building on the work of Barto, Singh, and Chentanez, addresses one of the central challenges in developmental robotics: how machines can autonomously discover and represent increasingly sophisticated skills through self-motivation. Stout also contributed to the practical deployment of social robots, organizing the influential “HRI 2018 Workshop” which examined the societal implications of affordable, mass-produced robot companions. Earlier in his career, he demonstrated his engineering acumen by leading Swarthmore College’s two-robot team, Frodo and Rose, to second and third place finishes in the 2002 AAAI Mobile Robot Competition’s Urban Search and Rescue and Robot Host events. His modular software architecture for heterogeneous robot tasks, described in that competition’s proceedings, remains a reference for multi-robot coordination. Stout’s work bridges theoretical advances in intrinsic motivation with tangible robotics applications, making him a notable figure in the drive toward autonomous, socially aware machines.