Modeling of Human Operator Behavior for Brain-Actuated Mobile Robots Steering.

Area of Science:

• Robotics
• Neuroscience
• Human-Computer Interaction

Background:

• Human operator control of brain-actuated robots using electroencephalograph (EEG) signals is complex and varies individually.
• Existing models lack the ability to decouple the user from the loop for improved system design and testing.

Purpose of the Study:

• To propose a novel operator brain-controlled steering model.
• To capture individual operator differences and brain-machine interface (BMI) performance variations.

Main Methods:

• Developed an operator decision model using the queuing network (QN) cognitive architecture.
• Created a new BMI performance model to represent accuracy in brain-controlled steering.
• Simulated and validated the model against human operator-in-the-loop experiments.

Main Results:

• The proposed QN-based operator decision model mimics human decision-making processes, accounting for individual differences.
• The BMI performance model effectively represents varying accuracy during brain-controlled operations.
• Model simulations successfully reproduced human operator behavior and direction control performance.

Conclusions:

• The developed operator brain-controlled steering model provides a valuable tool for system design and testing.
• The model's ability to capture individual differences and BMI performance enhances understanding of brain-actuated robot control.
• This approach facilitates the decoupling of the user from the loop, advancing the field of brain-actuated robotics.