Evaluation of Decision-making Ability in Car Driving While Operating Interior Devices Based on Probability-Weighted ARX Models

This paper proposes a technique for measuring the influence of in-car device operation on the decision-making ability of the driver. For several kinds of in-car devices such as air conditioner and navigation system, driving data are collected on a driving simulator where the driver follows the leading vehicle while operating the device. For each set of data, a mathematical model of driving behavior is identified as a Probability-Weighted ARX (PrARX) model. By calculating the model entropy of PrARX models, one can evaluate the influence of each device operation on the driver’s decision-making ability in a quantitative manner.