Double-layer staggered IWFBGs enable multi-dimensional tactile sensing on scalable flexible skin

Area of Science:

• Robotics and Materials Science
• Optoelectronics and Sensor Technology

Background:

• Existing tactile sensors face challenges like electromagnetic interference, complex wiring, and limited functionality.
• Dexterous robots and human-machine interfaces require advanced flexible tactile skins for comprehensive sensing.
• Optical tactile skins offer potential but often lack functionality or scalability.

Purpose of the Study:

• To develop a flexible optical tactile skin capable of simultaneous sensing of contact location, normal force, and contact orientation.
• To optimize the sensor design using a simulation-driven framework and improve its performance with advanced algorithms.
• To demonstrate the system's capability for spatially continuous tactile information prediction.

Main Methods:

• Utilized double-layer staggered identical weak fiber Bragg gratings (IWFBGs) interrogated by optical frequency-domain reflectometry (OFDR).
• Employed a simulation-driven design framework combining finite element analysis and neural networks to identify optimal IWFBG layout.
• Implemented a MAD-assisted moving-average detrending algorithm to mitigate temperature drift and laser instability.
• Applied a backpropagation (BP) neural network for tactile information prediction on a 70mm x 70mm sensor.

Main Results:

• Achieved localization accuracy with a root mean square error of 3.2-3.5 mm.
• Obtained force sensing with a mean absolute error of 0.27 N.
• Reached 88% directional accuracy across four contact directions.
• Demonstrated spatially continuous resolution capability through prediction on unseen locations.

Conclusions:

• The developed optical tactile skin offers a robust solution for multimodal tactile sensing in robotics and human-machine interfaces.
• The combination of IWFBGs, OFDR, and advanced algorithms provides high accuracy and continuous spatial resolution.
• This technology shows significant potential for advanced applications requiring dexterous manipulation and intuitive interaction.