Tactile Near-Sensor Analogue Computing for Ultrafast Responsive Artificial Skin.

Area of Science:

• Materials Science
• Robotics
• Neuroscience

Background:

• Current artificial skin systems suffer from high latency due to front-end electronics and analogue-to-digital conversions.
• This limits applications in prosthetics, robotics, and human-machine interactions requiring real-time tactile feedback.

Purpose of the Study:

• To develop an ultrafast and energy-efficient artificial skin system.
• To overcome the limitations of conventional artificial skin by integrating sensing and computation.

Main Methods:

• A flexible memristor array based on hafnium oxide was developed.
• This array was seamlessly integrated with a tactile sensor array.
• The system performs near-sensor analogue computing, eliminating the need for interface electronics.

Main Results:

• The system achieves sensing-computing operations in approximately 400 nanoseconds.
• It demonstrates real-time noise reduction and edge detection of tactile stimuli.
• Power consumption is reduced by an average of 1000 times compared to conventional systems.

Conclusions:

• Near-sensor analogue computing offers a viable pathway for ultrafast and energy-efficient artificial skin.
• This technology can enable next-generation tactile internet applications.
• The developed system shows potential for large-scale artificial skin implementations.