Flexible Magnetic Skin Sensor Array for Torsion Perception.

Area of Science:

• Biomedical Engineering
• Robotics
• Materials Science

Background:

• Current prosthetic hands lack the sophisticated tactile sensing of human hands, particularly in detecting torsional forces.
• Flexible tactile sensors offer a promising solution for replicating the natural stretch and movement of human skin in prosthetics.
• A significant gap exists in prosthetic technology regarding the sensation of torsional loads, crucial for fine motor control.

Purpose of the Study:

• To investigate the efficacy of a flexible magnetic sensor array coupled with an artificial neural network (ANN) for detecting and classifying torsional forces.
• To address the limitations in sensory feedback for upper limb prosthetics.

Main Methods:

• A flexible magnetic sensor array (3x3 magnets in elastomer over Hall effect sensors) was developed.
• A robotic arm applied ten distinct torque values to the sensor array for consistent data collection.
• An artificial neural network (ANN) was trained using the collected sensor data to classify applied torques.

Main Results:

• The ANN achieved an average training classification accuracy of 97.48% ± 0.33% in predicting applied torques.
• The flexible magnetic sensor successfully detected displacements in magnetic fields caused by applied torques.
• Consistent data acquisition was ensured through robotic arm application of torque.

Conclusions:

• The developed flexible magnetic sensor array and ANN system demonstrates high accuracy in detecting and classifying torsion.
• This novel sensor technology has the potential to significantly enhance tactile sensation in prosthetic hands.
• Improved sensory feedback could lead to more intuitive and functional prosthetic devices for amputees.