Inconsistency Calibrating Algorithms for Large Scale Piezoresistive Electronic Skin

Area of Science:

• Robotics and Materials Science
• Focuses on the intersection of advanced materials and robotic systems.

Background:

• Large-scale electronic skin is crucial for future human-robot interaction (HRI).
• Force-sensitive piezoresistive materials are key components for this technology.
• Material creep inconsistency poses a significant challenge to sensor accuracy.

Purpose of the Study:

• To present a non-array, large-scale piezoresistive tactile sensor.
• To develop novel calibration methods to enhance sensor accuracy and reliability.
• To address creep inconsistency in large-scale piezoresistive materials.

Main Methods:

• A creep tracking compensation method using K-means clustering and fuzzy pattern recognition was developed.
• Hierarchical clustering was used to classify and fuse regional piezoresistive properties.
• A Back-Propagation (BP) neural network model was constructed for force detection, incorporating position information.

Main Results:

• The proposed calibration methods effectively improved force detection.
• The creep tracking compensation significantly enhanced sensor accuracy by addressing material inconsistencies.
• The developed tactile sensor demonstrated reliable performance in detecting contact position and force.

Conclusions:

• The novel calibration techniques are effective for large-scale piezoresistive tactile sensors.
• The methods successfully improve detection accuracy in human-robot interaction applications.
• This work contributes to the advancement of reliable and accurate electronic skin technology.