Integrated Multifunctional Electronic Skins with Low-Coupling for Complicated and Accurate Human-Robot Collaboration

Area of Science:

• Robotics
• Materials Science
• Sensor Technology

Background:

• Advanced robot systems require sophisticated electronic skin (e-skin) for direct interaction with humans and environments.
• Low coupling and multifunctional e-skin are crucial for enhanced robotic capabilities.

Purpose of the Study:

• To develop a versatile e-skin system integrating multiple sensing capabilities.
• To demonstrate the e-skin's effectiveness in real-world robotic applications and improve performance with machine learning.

Main Methods:

• Vertical integration of diverse sensing materials and structures to create a multifunctional e-skin.
• Simultaneous sensing of proximity, pressure, temperature, and relative humidity.
• Implementation of machine learning for dynamic threshold adjustment to improve sensing of insulating objects.

Main Results:

• The e-skin achieved high sensitivity in sensing proximity (0.72 mm⁻¹), pressure (16.34 N⁻¹), temperature (0.0032 °C⁻¹), and humidity (15.2 pF/%RH).
• Cross-coupling errors were minimized (<2.65%) after temperature compensation.
• Machine learning significantly improved the accuracy of detecting insulating objects from 18% to 88%.

Conclusions:

• The developed multifunctional e-skin system offers robust sensing capabilities for advanced robotics.
• The system demonstrates accurate control and safety features in complex environments.
• This technology holds significant potential for human-machine collaboration and industrial safety.