Takahiro Hasegawa

Nara Institute of Science and Technology, Chubu University

Papers

Total Citations

H-Index

About

Takahiro Hasegawa is a robotics researcher specializing in dexterous manipulation, shared control, and grasp detection for robotic hands. His work bridges the gap between human teleoperation and autonomous robotic systems, with a focus on handling non-rigid objects and industrial parts. In his most-cited paper (2015, 4 citations), Hasegawa proposed a novel model-free reinforcement learning approach to learn shared control policies for dexterous telemanipulation, demonstrated through a page-turning skill—a challenging task requiring fine force and position control. He further advanced grasp detection with the Fast Graspability Evaluation (FGE) method (2019, 3 citations), which uses eigenvalue templates for fast and precise detection of object grasping positions, improving industrial robot efficiency. His work on deep convolutional neural networks for grasping detection (2018, 2 citations) integrates graspability metrics to enhance accuracy in everyday and industrial contexts. Though his citation counts are modest, Hasegawa’s contributions are foundational in practical robotic manipulation, offering scalable solutions for living-support and manufacturing robots. His research is particularly notable for combining learning-based approaches with real-time applicability, making him a key figure in advancing autonomous robotic dexterity.