Takayuki SOMEI
Papers
1
Total Citations
4
H-Index
1
About
Dr. Takayuki Somei is a robotics researcher whose work focuses on enabling autonomous systems to perceive and interact with their environments without relying on pre-programmed knowledge. His key research areas include robot perception, unsupervised learning, and embodied cognition. Somei’s most notable contribution is a novel framework for clustering image features based on the physical interactions—contact and occlusion—between a robot’s body and objects in its surroundings. By leveraging statistical dependencies and conditional probability, his approach allows robots to autonomously segment and recognize visual features from raw sensory data, a foundational step toward more adaptive and self-learning machines. Though his most-cited paper, published in 2013, has garnered 4 citations, its conceptual impact lies in bridging low-level sensorimotor data with higher-level perceptual organization—a challenge central to developmental robotics. Somei’s work offers a principled method for robots to build internal models of their world through active exploration, contributing to the broader goal of creating machines that learn like biological organisms. His research remains a touchstone for those studying unsupervised feature learning in robotics.
Research Focus
Key Achievements
Top Papers
- 1