Corey Goldfeder

Columbia University, Google (United States)

Papers

Total Citations

557

H-Index

About

Corey Goldfeder is a leading researcher in robotic grasping and manipulation, whose work bridges computational geometry, neuroscience, and practical robotics. His primary contributions lie in developing data-driven and dimensionality-reduction approaches to enable more efficient and stable robotic grasping. Goldfeder’s seminal paper, “Grasp Planning via Decomposition Trees” (2007, 219 citations), introduced a planner that accounts for the relative sizes of robotic hands and objects, overcoming physical joint and positioning limitations to generate realizable grasps. In parallel, his work “Dimensionality Reduction for Hand-Independent Dexterous Robotic Grasping” (2007, 218 citations) applied insights from neuroscience—showing that human hand control operates in a low-dimensional configuration space—to robotic hands, dramatically simplifying grasp planning. This concept was further expanded in “Data-driven Grasping” (2011, 114 citations), which leveraged large datasets to generalize grasping strategies across diverse objects. By reducing the complexity of high-dimensional hand configurations, Goldfeder’s research has made dexterous manipulation more computationally tractable, influencing both autonomous robotics and prosthetic design. His work remains foundational for researchers seeking to bridge the gap between human-like dexterity and robotic efficiency.