SEOUL -- Inspired by the frequency tunability of the inner structure of the cochlea in the human ear, various frequency-selective acoustic sensors have been created, but developments have been limited by the narrow tunable range of resonance frequencies and difficulties in miniaturization. South Korean researchers have developed frequency-selective acoustic and haptic sensors for dual-mode human-machine interfaces based on triboelectric sensors.
Developments in human-machine interfaces requiring high-frequency signal detection for robots, virtual reality, augmented reality, and Internet of Things (IoT) demand precise and intuitive control to deliver various senses and biosignals from humans. The accurate transmission of biosignals without the interference of surrounding noises is a key factor for the realization of human-machine interfaces.
Frequency-selective acoustic and haptic sensors developed by a research team from the Ulsan National Institute of Science and Technology (UNIST) showed high sensitivity and linearity under a wide range of dynamic pressures and resonance frequency, which enables high acoustic frequency selectivity in a wide frequency range from 145 to 9000 Hz,
The frequency-selective multichannel acoustic sensor array combined with an artificial neural network demonstrates over 95 percent accurate voice recognition for different frequency noises ranging from 100 to 8000 Hz, the research team said in a paper published on the website of Science Advances, a peer-reviewed multidisciplinary open-access scientific journal.
The dual-mode sensor with linear response and frequency selectivity over a wide range of dynamic pressures facilitated the differentiation of surface texture and control of an avatar robot using both acoustic and mechanical inputs without interference from surrounding noise, the team said.
The development of dynamic human-machine interfaces requires the selective recognition of desired frequency information without interference from surrounding noises. Researchers found that triboelectric sensors instantly generated high power in response to dynamic stimuli without additional power supply and recognized the multiple physical touch, motion, fine texture, and displacement of objects.
The UNIST team proposed a frequency-selective acoustic and haptic smart glove for dual-mode human-machine interfaces, as well as a hierarchical ferroelectric composite comprising surface macrodome and inner micropore structures decorated with nanoparticles to develop a self-powered frequency-selective triboelectric sensor with high sensitivity and linear response.
"We further demonstrated a triboelectric sensor-based smart glove for human-machine interfaces to remotely control various motions of robotic hands," the team said, adding that the capabilities of hierarchical triboelectric sensors based on ferroelectric composites provide a solid platform for improving conventional sensors and their application in humanoid robots, wearable devices and biometrics.