Driver Cognitive Distraction Recognition Based on Multi-Source Data from Simulated Driving Experiments

Nowadays, cognitive distraction in the process of driving has become a frequent phenomenon, which has led to a certain proportion of traffic accidents, causing a lot of property losses and casualties. Since the fact that cognitive distraction is mostly reflected in the driver's reception and thinking of information unrelated to driving, it is difficult to recognize it from the driver's facial features. As a result, the accuracy of prediction is usually lower relying solely on facial performance to detect cognitive distraction. In this research, fifty participants took part in our simulated driving experiment. And each participant conducted the experiment in four different traffic scenarios using a high-fidelity driving simulator, including three cognitive distraction scenarios and one normal driving scenarios. Firstly, we identified the facial performance indicators and vehicle performance indicators that had a significant effect on cognitive distraction through one-way ANOVA. Then we applied the YOLOv5s model to detect cognitive distraction by combining the above two types of performance indicators. The results showed that in deep learning models, the accuracy of detecting driver cognitive distraction by combining the facial data and the vehicle data was 61.69%, and the recall rate was 83.28%, which were 8.91% and 15.1% higher than the ones using only the facial data.