Driver distraction judging model under in-vehicle information system operation based on driving performance

In order to explore the recognition method of distracted driving in operating in-vehicle information system (IVIS), the effectiveness of driving performance indicator as the evaluation variable of distracted driving was verified by using variance analysis method. Experiments under IVIS operation based on driving performance were carried out, and road scene video and driving performance indicators data were obtained. According to the driving performance data under IVIS operation, using support vector machine (SVM) classification algorithm, the driver distraction judging model was built based on vehicle driving performance under IVIS operation, and the effectiveness of the model was validated by experimental data. The results show that SVM model can be used to determine the driver distraction state. When using RBF kernel function, the recognition accuracy of driving distraction is 89. 86%, higher than those by using Sigmoid and polynomial kernel function. The model can effectively determine the driver distraction state, which can provide data support for driving distraction control strategy.