Daisuke Shinohara

About

Daisuke Shinohara is a robotics researcher whose work focuses on the intersection of machine learning and physical interaction, specifically teaching anthropomorphic robots to handle non-rigid materials. His primary research areas include reinforcement learning, motor skill acquisition, and robotic manipulation of deformable objects. Shinohara’s major contribution lies in developing novel frameworks that enable robots to learn complex, real-world tasks involving flexible materials—such as wearing a T-shirt, turning socks inside out, or applying bandages—which are notoriously difficult for traditional rigid-control systems. His most cited paper, "Reinforcement learning of a motor skill for wearing a T-shirt using topology coordinates" (2013, 42 citations), introduces a topology-based coordinate system that allows robots to quantitatively define and learn these intricate motor skills through trial and error. This work, along with his earlier foundational study "Learning motor skills with non-rigid materials by reinforcement learning" (2011, 9 citations), has laid important groundwork for assistive robotics and automated caregiving. Shinohara’s research is particularly notable for its practical implications in healthcare and elderly care, where robots capable of dressing or bandaging patients could significantly improve quality of life.

Research Focus

Key Achievements

Top Papers

1. 1
   Reinforcement learning of a motor skill for wearing a T-shirt using topology coordinates

   42 citations · 2013
2. 2
   Learning motor skills with non-rigid materials by reinforcement learning

   9 citations · 2011

Key Collaborators

Contact & Links