Hypovigilance detection for UCAV operators based on a hidden Markov model

Area of Science:

• Aerospace Engineering
• Cognitive Neuroscience
• Human Factors Engineering

Background:

• Unmanned Combat Aerial Vehicles (UCAVs) are increasingly utilized in modern military operations.
• UCAVs exhibit a significantly higher accident rate compared to manned aircraft.
• Operator hypovigilance, a decrease in vigilance during operation, is a primary contributor to UCAV accidents.

Purpose of the Study:

• To propose and evaluate novel hypovigilance detection models for UCAV operators.
• To mitigate the risk of accidents by identifying and addressing operator hypovigilance.
• To enhance the safety and effectiveness of UCAV operations.

Main Methods:

• Development of hypovigilance detection models utilizing electroencephalography (EEG) signals.
• Application of Hidden Markov Models (HMMs) to differentiate between normal vigilance and hypovigilance states.
• Training operator-specific HMMs as individual detection models.
• Validation through real-world experiments using EEG data acquisition devices.

Main Results:

• The proposed HMM-based models demonstrated satisfactory efficacy and effectiveness in detecting operator hypovigilance.
• Experimental results confirmed the models' capability to accurately identify vigilance state changes.
• The models provide a reliable method for monitoring UCAV operator cognitive states.

Conclusions:

• The developed EEG-based hypovigilance detection models offer a promising solution to reduce UCAV accident rates.
• Implementing these models can significantly improve the safety and reliability of unmanned aerial vehicle operations.
• Addressing operator hypovigilance is crucial for the future advancement of autonomous and remotely piloted systems.