An Efficient Multi-Channel Electrotactile Parameter Configuration Method for Personalized Teleoperation

Area of Science:

• Robotics and Human-Computer Interaction
• Biomedical Engineering
• Computational Neuroscience

Background:

• Electrotactile feedback offers compact tactile cues for robotic teleoperation.
• Personalized calibration of electrotactile systems is time-consuming due to inter-user variability in tactile perception.

Purpose of the Study:

• To develop a subject-informed computational model for optimizing electrotactile feedback configurations.
• To propose a simulation-derived ranking method, the Perceived Correctness Score (PCS), to expedite personalized calibration.

Main Methods:

• A multi-layer finite element model of fingertip electric-field distribution was coupled with a neural-response model.
• A gradient boosting regression model recommended optimal electrode diameter-spacing combinations from 36 candidates.
• Validation involved a shape/area recognition task with six subjects using a custom electrotactile system.

Main Results:

• The predicted PCS moderately correlated with measured recognition accuracy (Pearson r = 0.48, p < 0.05).
• The model achieved Top-1 exact matching for 50% of subjects and Top-5 coverage for 83%.
• Average configuration time was reduced by 87.0%, from 122.7 min to 16.0 min.

Conclusions:

• Model-guided ranking substantially reduces the time and effort required for individualized electrotactile configuration.
• This approach enhances the efficiency of deploying personalized tactile feedback systems in robotic applications.