Related Experiment Videos
Driver's mental workload prediction model based on physiological indices
Shengyuan Yan1, Cong Chi Tran1,2, Yingying Wei1
1a College of Mechanical and Electrical Engineering , Harbin Engineering University , China.
Summary
This study developed a predictive model for new drivers' mental workload (MWL) using physiological data and subjective ratings. The model accurately predicts driving errors, aiding in performance improvement.
Area of Science:
- Human-Computer Interaction
- Cognitive Psychology
- Transportation Safety
Background:
- Driver mental workload (MWL) is critical for performance, especially for novice drivers.
- Predicting MWL can enhance driving safety and training effectiveness.
- Current methods for assessing MWL may not be fully integrated or predictive.
Purpose of the Study:
- To investigate the correlation between new drivers' MWL and their driving performance (number of errors).
- To develop a predictive model for driver MWL using subjective and physiological data.
- To establish a reference for new drivers' MWL and inform personalized driving lesson plans.
Main Methods:
- Utilized the group method of data handling to construct the predictive model.
- Incorporated subjective workload ratings via the NASA task load index (NASA-TLX).
- Integrated six physiological indices alongside NASA-TLX for MWL assessment.
Main Results:
- A significant positive correlation was found between NASA-TLX scores and the number of driving errors.
- The developed predictive model demonstrated strong validity with an R-squared value of 0.745.
- The model successfully linked physiological indices to predicted driver MWL.
Conclusions:
- The proposed model provides a valid method for predicting new driver MWL and performance.
- Physiological indices can serve as a reference for assessing new drivers' mental workload.
- The model supports the development of targeted driving lesson plans to optimize MWL and performance.