Driving assistant using generative AI pre-generated messages in simulator-based driving assessment: A step towards...

Area of Science:

• Human-Computer Interaction
• Transportation Safety
• Psychology

Background:

• Traditional on-road driving assessments are risky and costly.
• High-cost simulator systems limit accessibility for research.
• There is a need for affordable, accessible driving assessment tools.

Purpose of the Study:

• To develop a low-cost simulator-based driving assessment system.
• To integrate a Generative AI-powered speech assistant for real-time interaction.
• To enhance drivers' situational awareness and emotional states during assessment.

Main Methods:

• Developed a novel, low-cost simulator-based driving assessment system.
• Incorporated a speech-based assistant utilizing Generative AI for pre-generated messages.
• Assessed participants to evaluate driving performance and emotional states.

Main Results:

• Preliminary outcomes show encouraging progress in driving performance.
• Participants exhibited improved emotional states (high valence, medium arousal) with the assistant.
• The system demonstrated potential in identifying subpar performers.

Conclusions:

• The proposed system offers a viable, low-cost alternative for driving assessment.
• The speech-based assistant effectively enhances driver engagement and performance.
• This approach can improve the accessibility and effectiveness of driving evaluations.