ABSTRACT
Research indicates that monotonous automated driving increases the incidence of fatigued driving. Although many prediction models based on advanced machine learning techniques were proposed to monitor driver fatigue, especially in manual driving, little is known about how these black-box machine learning models work. In this paper, we proposed a combination of Gaussian Process Boosting (GPBoost) and SHapley Additive exPlanations (SHAP) to predict driver fatigue with explanations. First, in order to obtain the ground truth of driver fatigue, we used PERCLOS (percentage of eyelid closure over the pupil over time) between 0 and 100 as the response variable. Second, we built a driver fatigue regression model using both physiological and behavioral measures with GPBoost that was able to address the within-subjects correlations. This model outperformed other selected machine learning models with root-mean-squared error (RMSE) = 2.965, mean absolute error (MAE) = 1.407, and adjusted . Third, we employed SHAP to identify the most important predictor variables and uncovered the black-box GPBoost model by showing the main effects of the most important predictor variables globally and explaining individual predictions locally. Such an explainable driver fatigue prediction model offered insights into how to intervene in automated driving when necessary, such as during the takeover transition period from automated driving to manual driving.
Acknowledgments
This work was supported by Ford Summer Sabbatical Program.
Additional information
Notes on contributors
Feng Zhou
Feng Zhou received his Ph.D. degrees in Mechanical Engineering and Human Factors from Gatech and Nanyang Technological University in 2014 and 2011, respectively. He is currently Assistant Professor at the Department of Industrial and Manufacturing Systems Engineering, University of Michigan-Dearborn, focusing on human factors, engineering design, and machine learning.
Areen Alsaid
Areen Alsaid received the Ph.D. degree from the Industrial and Systems Engineering Department at the University of Wisconsin-Madison, Madison, Wi in 2020. She is currently a research scientist at Ford Motor Company. Her research interests include affective computing, driver state estimation, and driving safety.
Mike Blommer
Mike Blommer received his Ph.D. degree in Electrical Engineering from the University of Michigan, Ann Arbor in 1995, specializing in signal processing and psychoacoustics. He is currently a Technical Leader at Ford Motor Company, with research interests in driver warning systems, automated driving, HMI, and vehicle dynamics.
Reates Curry
Reates Curry received the Ph.D. degree in Biomedical Engineering from Rutgers University, NJ. She has been with the Ford Research & Innovation Center since 1995 working with the driving simulator team. Her expertise includes driver safety, eye tracking research, etc.
Radhakrishnan Swaminathan
Radhakrishnan Swaminathan received the Ph.D. degree in Electrical and Computer Engineering from Oakland University, MI, USA, in 2012. He is currently a Research Engineer with Research and Advanced Engineering, Ford Motor Company. He conducts research on driver HMI, distraction, driver warning systems, and automated driving in Ford’s VIRTTEX Driving Simulator.
Dev Kochhar
Dev Kochhar received his Ph.D degree in System Design from the University of Waterloo, Canada, in 1974. He was a Professor of industrial engineering (12 years), and a Technical Staff at AT&T Bell Laboratories (6 years). He is currently a retired Technical Specialist with the Ford Motor Company.
Walter Talamonti
Walter Talamonti received the Ph.D. degree in Industrial Engineering from Wayne State University, Detroit, MI, USA, in 2017. He is currently a Research Engineer with Research and Advanced Engineering, Ford Motor Company. He conducts human factors research in the areas of human–machine interface, driver warning systems, and autonomous driving.
Louis Tijerina
Louis Tijerina received the Ph. D. degree in Experimental Psychology from The Ohio State University in 1989. He is currently a retired Senior Technical Specialist at Ford Motor Company. His research interests include driver HMI, distraction, driver warning systems, and automated driving.