Papers
6
Total Citations
87
H-Index
6
About
Rok Vuga is a robotics researcher whose work sits at the intersection of human motion analysis, robot learning, and autonomous skill acquisition. His research focuses on enabling robots to learn complex, dynamically stable movements by combining insights from human motion capture with advanced machine learning techniques. Vuga’s most cited work, "Motion capture and reinforcement learning of dynamically stable humanoid movement primitives" (27 citations), addresses the fundamental challenge of transferring human motion to humanoid robots, which have different kinematics and dynamics. He developed a system using low-cost RGB-D cameras to capture human movements and convert them into stable robot trajectories. Vuga also pioneered the concept of Semantic Event Chains (SECs) for action representation, allowing robots to encode complex manipulations from 3D image streams in a generalizable way (18 citations). His integrated approach to policy learning, which fuses iterative learning control with reinforcement learning, demonstrates fast convergence and enhanced adaptation. With over 87 total citations across his key publications, Vuga’s contributions to structural bootstrapping, statistical generalization, and trajectory representation have advanced the field of cognitive robotics, making his work essential reading for researchers interested in autonomous robot learning and sensorimotor skill acquisition.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Exploiting previous experience to constrain robot sensorimotor learning18 citations · 2011
- 4
- 5Enhanced Policy Adaptation Through Directed Explorative Learning8 citations · 2015
- 6