Bioinspired Young's Modulus-Hierarchical E-Skin with Decoupling Multimodality and Neuromorphic Encoding Outputs to...

Area of Science:

• Biomaterials Science
• Neuroprosthetics
• Sensory Systems Engineering

Background:

• Restoring natural sensory feedback in prosthetics is crucial for user embodiment and function.
• Current prosthetic interfaces struggle to unambiguously transmit multimodal sensory information to the nervous system.

Purpose of the Study:

• To develop a bioinspired electronic skin capable of decoupling temperature and pressure sensing for enhanced prosthetic functionality.
• To create a neural coding and machine learning system for interpreting sensory signals and enabling object recognition.

Main Methods:

• Fabrication of a hierarchical structure MXene-based temperature-pressure electronic skin (TPD e-skin) inspired by human skin's modulus.
• Implementation of bionic microstructures and contact resistance modulation for sensitive and decoupled sensing.
• Development of a neural model for coding sensory data into nerve-acceptable frequency signals and a machine learning algorithm for signal fusion and object recognition.

Main Results:

• The TPD e-skin demonstrated high sensitivity across a wide pressure range with excellent temperature insensitivity (91.2% reduction).
• The device achieved pressure insensitivity of the thermistor due to its structural configuration.
• Neural coding successfully translated signals into three distinct frequency types, enabling four operational states.
• Object recognition accuracy reached 98.7% using a brain-like machine learning fusion process.

Conclusions:

• The developed TPD e-skin and neural system offer a promising approach for advanced prosthetic devices.
• This technology enables multimodality-decoupling sensing and deep neural integration, paving the way for more natural prosthetic functionality.