Predicting Driver Fatigue in Monotonous Automated Driving with Explanation using GPBoost and SHAP

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 R2=0.996. 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.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 42 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.