Papers
97
Total Citations
2,556
H-Index
27
About
Roderic A. Grupen is a pioneering roboticist whose research spans robot learning, motion planning, grasping, and autonomous skill acquisition. Based at the University of Massachusetts Amherst, Grupen has made foundational contributions to how robots perceive, plan, and learn in unstructured environments. His early landmark work on harmonic functions for path planning (1993, 288 citations) demonstrated that solutions to Laplace's equation could eliminate the spurious local minima that plagued classical potential-field methods — a contribution still widely cited in robotics curriculum today. Grupen subsequently turned to the challenge of robot grasping, developing nullspace control frameworks that allow robots to grasp objects of unknown geometry through active sensorimotor strategies rather than relying on precise geometric models, work spanning multiple highly cited papers through the 2000s. Perhaps his most recognized recent contribution is the Constructive Skill Trees (CST) algorithm (2011, 300 citations), which enables robots to learn reusable, hierarchically organized skills directly from demonstration trajectories. This work, alongside his research on autonomous skill acquisition and generalizable control programs, positions Grupen at the forefront of developmental and lifelong robot learning. His uBot platform further exemplifies his commitment to building capable, dynamically stable humanoid systems for real-world manipulation research.
Research Focus
Key Achievements
Top Papers
- 1Robot learning from demonstration by constructing skill trees300 citations · 2011
- 2The applications of harmonic functions to robotics288 citations · 1993
- 3
- 4Nullspace composition of control laws for grasping75 citations · 2003
- 5Null-Space Grasp Control: Theory and Experiments75 citations · 2010
- 6A feedback control structure for on-line learning tasks64 citations · 1997
- 7Autonomous Skill Acquisition on a Mobile Manipulator64 citations · 2011
- 8Designing a Self-Stabilizing Robot for Dynamic Mobile Manipulation59 citations · 2006
- 9A hybrid architecture for adaptive robot control58 citations · 2000
- 10Learning Generalizable Control Programs57 citations · 2011