Guodong Shi

Papers

Total Citations

H-Index

About

Guodong Shi is a researcher whose work spans the intersecting frontiers of robotics, multi-agent systems, and control theory. His contributions reflect a broad technical vision that bridges theoretical rigor with practical application in intelligent autonomous systems. In robotic manipulation, Shi co-developed the Fast-Learning Grasping (FLG) framework, a deep reinforcement learning approach that enables robots to efficiently handle cluttered environments through integrated pre-grasping actions such as pushing and shifting — a contribution that has already garnered 10 citations since its 2021 publication. His theoretical work extends into distributed intelligence, as demonstrated by his monograph on multi-agent online optimization, which offers a comprehensive treatment of how networked agents can make robust decisions under sequentially arriving, uncertain information — a foundational problem in modern autonomous systems. More recently, Shi has pushed into the challenging domain of visual-inertial SLAM, introducing the PEBO-SLAM framework, a principled observer design that provides formal convergence guarantees for simultaneous localization and mapping on nonlinear manifolds. Together, these contributions position Shi as a researcher deeply committed to developing theoretically grounded yet deployable solutions at the frontier of autonomous robotics and distributed decision-making.