Tactile sensing-based control algorithm for real-time grasp synthesis in object manipulation tasks of humanoid robot fingers

This paper presents development of tactile sensing-based control algorithm for humanoid robot finger system with optical three-axis tactile sensor mounted on fingertips. Our aim is to develop an intelligent control system that can recognize stiffness of unknown objects and respond to sudden changes of objectpsilas weight during object manipulation. For this purpose, we developed a novel optical three-axis tactile sensor system based on an optical waveguide transduction method capable of acquiring normal and shearing forces. We proposed a control algorithm in the finger control system based on tactile and slippage sensations, and analyzed real-time grasp synthesis in object manipulation tasks. The control algorithm was designed to control fingertips movements by defining optimum grasp pressure and perform re-push movement when slippage was detected in object manipulation tasks. Verification experiments using humanoid robot fingers were conducted whose results revealed that the fingerpsilas system managed to recognize the stiffness of unknown objects and complied with sudden changes of the objectpsilas weight during object manipulation tasks.