Haptic in-sensor computing device made of carbon nanotube-polydimethylsiloxane nanocomposites

Kouki Kimizuka,Saman Azhari,Shoshi Tokuno,Ahmet Karacali,Yuki Usami,Shuhei Ikemoto,Hakaru Tamukoh,Hirofumi Tanaka
2024-06-06
Abstract:The importance of haptic in-sensor computing devices has been increasing. In this study, we successfully fabricated a haptic sensor with a hierarchical structure via the sacrificial template method, using carbon nanotubes-polydimethylsiloxane (CNTs-PDMS) nanocomposites for in-sensor computing applications. The CNTs-PDMS nanocomposite sensors, with different sensitivities, were obtained by varying the amount of CNTs. We transformed the input stimuli into higher-dimensional information, enabling a new path for the CNTs-PDMS nanocomposite application, which was implemented on a robotic hand as an in-sensor computing device by applying a reservoir computing paradigm. The nonlinear output data obtained from the sensors were trained using linear regression and used to classify nine different objects used in everyday life with an object recognition accuracy of >80 % for each object. This approach could enable tactile sensation in robots while reducing the computational cost.
Emerging Technologies,Neural and Evolutionary Computing
What problem does this paper attempt to address?