Precise perception of surface tackiness enabled by a soft single-sensing-element tactile sensor.

Area of Science:

• Robotics
• Materials Science
• Sensor Technology

Background:

• Precise tactile sensing is crucial for robots interacting with complex environments.
• Measuring surface tackiness requires simultaneous, decoupled measurement of pressure and pulling forces.
• Existing sensors face challenges like signal crosstalk and baseline instability.

Purpose of the Study:

• To develop a tactile sensor capable of intrinsically decoupled measurement of pressure and pulling forces for adhesion analysis.
• To create a sensor that overcomes the limitations of existing technologies in measuring tackiness.
• To enhance robotic interactive intelligence for handling delicate and tacky objects.

Main Methods:

• A surface-soft, magneto-mechanical coupling tactile sensor with a skin-like bidirectional deformation design was engineered.
• Inward pressure and outward pulling forces were designed to generate baseline-separated magnetic signatures.
• The sensor's performance was evaluated for stability, signal coincidence, and accuracy in tackiness identification, including integration with a neural network.

Main Results:

• The sensor achieved intrinsic signal decoupling within a single sensing element, eliminating the need for complex post-processing.
• Demonstrated high stability with only 0.25% force drift over 10 hours and minimal drift after mechanical disturbances.
• Achieved 99.52% signal coincidence across repeated press-pull cycles and 99.78% tackiness identification accuracy when integrated with a neural network.

Conclusions:

• The developed tactile sensor provides continuous, high-stability monitoring of the full adhesion cycle.
• Its performance rivals standard adhesion testing, enabling precise tackiness differentiation.
• This technology lays a foundation for advanced robotic manipulation of tacky and lightweight objects, surpassing human precision.