Bioinspired kinesthetic system for human-machine interaction

As the most basic and important sense of the human body, kinesthesia has the ability to act and adapt to stimuli. Simulating the kinesthetic process from the level of sensory neurons is an important task toward the emulation of neuromorphic computation, while currently report for artificial kinesthetic system is still not available. Hence, in this work, an artificial kinesthetic system is developed which consists of a single-electrode triboelectric nanogenerator (S-TENG) that can be attached to human skin and a field effect synaptic transistor (FEST). The S-TENG based on PDMS/MXene friction layer exhibits high sensitivity of 0.197 kPa(-1) in a low-pressure region (<6 kPa) and 0.003 kPa(-1) in a high-pressure region (6-30 kPa). FEST achieves synaptic plasticity in biology and simulates the short-term to long-term memory transition and learning process. Artificial kinesthetic system can readily achieve the perception of human muscle/joint motion state and orientation information. In addition, the assessment of fatigue driving risk is realized which substantially improves the efficiency and accuracy of the instruction recognition process. Furthermore, the identification of ASL (American Sign Language) gestures is simulated to demonstrate the recognition accuracy. This work shows a widespread potential in the construction of next-generation neuromorphic sensory network, neurorobotics and interactive artificial intelligence.

Automated summary

Kinesthesia, as a fundamental human sense, plays a vital role in responding and adapting to stimuli. This study introduces an artificial kinesthetic system that can perceive muscle/joint motion and orientation, while also assessing fatigue driving risk for improved efficiency in instruction recognition.