A Versatile Robotic Hand for Garment Handling in Manufacturing

Speaker

Jose Victorio Salazar Luces, Tohoku University

Co-author

Xinhao Chen, Eichi Machida

Abstract

Traditional cloth manipulation research often employs grippers with low degrees of freedom (DOF) to simplify hardware. However, this approach creates significant challenges for the algorithms needed to perceive and estimate the state of the cloth, due to the complex and non-linear dynamics of fabric behavior. To address these limitations, we propose a novel approach that combines a grasping mechanism with a suction system. This combined approach enables the robot hand to perform complex tasks, traditionally requiring hands with many DOFs and intricate control, while circumventing the disadvantages caused by simplifying the mechanism and control complexity. On this talk we are going to describe the design of a underactuated robot hand with a suction system and show its effectiveness through various tasks.

Speaker Bio

Jose Salazar is an Assistant Professor at Tohoku University, Japan. He obtained his Bachelor degree in Systems Engineering at Los Andes University, in Merida, Venezuela. In 2013, he was awarded the MEXT Scholarship to pursue studies in robotics at Tohoku University. He obtained his Master’s and Ph.D. at Tohoku, and started working at the Smart Robots Design Laboratory from 2018. From October 2021 to April 2022, Jose was appointed as a Visiting Assistant Professor in the Department of Electrical and Electronic Engineering at The University of Hong Kong. He is currently working as an area expert on a joint project between HKU and Tohoku University.