Sanket Kamthe
Papers
2
Total Citations
310
H-Index
2
About
Sanket Kamthe is a leading researcher at the intersection of robotics, reinforcement learning, and human-robot interaction. His work is distinguished by pioneering methods that enable robots to learn complex cooperative behaviors with remarkable data efficiency. Kamthe’s seminal paper, “Interaction Primitives for Human-Robot Cooperation Tasks” (2014), with 200 citations, introduced a foundational framework for robots to anticipate and adapt to human partners’ movements in real time, moving beyond rigid, pre-programmed sequences. This work has been highly influential in making human-robot collaboration more fluid and intuitive. Building on this, Kamthe’s highly cited “Data-Efficient Reinforcement Learning with Probabilistic Model Predictive Control” (2017, 110 citations) directly tackled a critical bottleneck in robotics: the need for vast amounts of trial-and-error data. By fusing model predictive control with probabilistic inference, he developed an algorithm that learns optimal policies from far fewer interactions, making RL practical for real-world physical systems where every failure carries a cost. Kamthe’s contributions are essential for creating autonomous robots that can safely and skillfully work alongside people in manufacturing, healthcare, and domestic settings.
Research Focus
Key Achievements
Top Papers
- 1Interaction primitives for human-robot cooperation tasks200 citations · 2014
- 2