Jared Shamwell

University of Maryland, College Park, Draper Laboratory

Papers

Total Citations

H-Index

About

Jared Shamwell is a robotics researcher whose work bridges machine learning, metacognition, and human-robot interaction. His most influential contributions center on the "Robot Baby" paradigm, where he envisions robots as developmental learners that use massive metacognition—a synergistic interplay between machine learning and commonsense reasoning—to understand their environments. In his foundational 2012 papers, Shamwell proposed long-term projects where mobile robots, like infants, gradually build knowledge about their surroundings, with early steps implemented through growing neural gas networks. These works, accumulating 9 and 8 citations respectively, laid the groundwork for more adaptive robotic learning systems. Shamwell also contributed a comprehensive 2013 review bridging reinforcement learning and robotics, comparing algorithms like Q-learning and Actor-Critic methods, which has garnered 6 citations. His 2016 work on grounded self-symbols for human-robot interaction explores how robots can develop self-awareness to improve collaboration with humans. Most recently, in 2024, Shamwell has advanced deep uncertainty modeling for robotic state estimation, addressing non-Gaussian aleatoric uncertainty—a critical step toward more reliable perception in real-world robotics. His research trajectory reflects a persistent focus on making robots more autonomous, self-aware, and capable of lifelong learning.