Patricio Rivera
Papers
2
Total Citations
19
H-Index
2
About
Patricio Rivera is a leading researcher in robotic manipulation, with a primary focus on enabling dexterous, human-like control of anthropomorphic robotic hands. His work sits at the intersection of deep reinforcement learning, robotics, and biomechanics. Rivera’s major contribution is pioneering the use of a “synergy space” derived from natural hand poses to dramatically simplify the high-dimensional control problem of multi-fingered hands. By embedding these biologically-inspired priors into RL frameworks, he has shown that robots can learn complex, stable object manipulation tasks that would otherwise be intractable. His 2021 paper on this synergy-space approach has garnered 15 citations, establishing a foundational method in the field. In subsequent work, Rivera tackled the critical challenge of reward engineering, developing reward-shaping techniques that leverage hand pose priors to enable more reliable and sample-efficient on-policy learning. His research directly addresses the core difficulty of transferring human dexterity to machines, with implications for prosthetics, manufacturing, and assistive robotics. Rivera’s work is distinguished by its elegant fusion of biological insight with cutting-edge machine learning, making him a key voice in the quest for truly anthropomorphic robotic manipulation.
Research Focus
Key Achievements
Top Papers
- 1
- 2