Yusuke Bansho
Papers
1
Total Citations
111
H-Index
1
About
Yusuke Bansho is a researcher whose work sits at the intersection of robotics, tactile sensing, and deep learning. His most significant contribution lies in advancing tactile object recognition, a critical capability for robots operating in environments where vision is unreliable or unavailable. In his highly cited 2014 paper, "Tactile object recognition using deep learning and dropout" (111 citations), Bansho pioneered the use of deep neural networks—specifically employing the dropout regularization technique—to recognize objects through power grasping, even when their orientation and position relative to the robotic hand are unknown. This work demonstrated a robust, multimodal approach to tactile sensing, enabling robots to identify objects purely through touch. By addressing the challenge of recognizing objects with arbitrary poses, Bansho’s research has had a notable impact on the development of more dexterous and perceptive robotic hands. His contributions are particularly relevant for applications in manufacturing, assistive robotics, and autonomous manipulation, where tactile feedback is essential for safe and effective interaction with the physical world.
Research Focus
Key Achievements
Top Papers
- 1Tactile object recognition using deep learning and dropout111 citations · 2014