SmartCar: Detecting Driver Stress

Smart physiological sensors embedded in an automobile aaord a novel opportunity to capture naturally occurring episodes of driver stress. In a series of ten ninety minute drives on public roads and highways, electrocardiogram, elec-tromyogram, respiration and skin conductance sensors were used to measure autonomic nervous system activation. The signals were digitized in real time and stored on the SmartCar's pen-tium class computer. Each drive followed a pre-speciied route through fteen diierent events, from which four stress level categories were created according to the results of the subjects self report questionnaires. In total, 545 one minute segments were classiied. A linear discriminant function was used to rank each feature individually based on recognition performance and a sequential forward oating selection (SFFS) algorithm was used to nd an optimal set of features for recognizing patterns of driver stress (88.6%). Using multiple features improved performance signiicantly over the best single feature performance (62.2%).