Shawn Squire

University of Maryland, Baltimore County, University of Maryland, College Park

Papers

Total Citations

114

H-Index

About

Shawn Squire’s research lies at the intersection of robotics, natural language processing, and hierarchical reinforcement learning, with a focus on making autonomous systems more intuitive and efficient. His most influential work, “Grounding English Commands to Reward Functions” (65 citations), pioneered a method for translating natural language instructions into reward functions through demonstration, enabling non-experts to program robots without technical expertise. This breakthrough addresses a critical barrier to human-robot collaboration. Squire further advanced the field with “Planning with Abstract Markov Decision Processes” (49 citations), which introduced hierarchical abstractions for planning under uncertainty in large state-action spaces. This work enables robots to efficiently adapt to changing reward functions in complex, human-scale environments—a fundamental challenge for real-world deployment. By combining language grounding with scalable planning, Squire’s contributions bridge the gap between accessible human-robot interaction and computationally tractable decision-making. His research has been cited over 114 times, reflecting its impact on both theoretical frameworks and practical robotics applications. Squire’s work continues to influence how robots understand commands and plan intelligently in dynamic settings.