ABSTRACT
ABSTRACT: The driver state information is indispensible for the development and evaluation of an advanced driver assistance system. To propose a driver mental workload evaluation method, physiological data, and subjective scores were recorded in a moving-base driving simulator under a dual-task condition composed of a driving task and an auditory n-back task. The heart rate variability, skin conductance level, and respiration rates were extracted and examined as a measure of a driver’s workload. Subjective scores used the NASA-TLX and indicated that the driver’s stress level increases under the dual-task condition, and a significant variation of physiological indices were found in the experimental trials. The results indicated a correlation between the physiological indices and the driver’s mental workload. In order to verify the effectiveness of a simulator study, a combined measure was then created, using a multiple regression method based on the physiological indices. The evaluation method developed in this study can be used in the design of an advanced driver assistance system and a human-machine interface.
