Search Results

Drivers often engage in secondary in-vehicle activity that is not related to vehicle control. This may be functional and/or to relieve monotony. Regardless, drivers believe they can safely do so when their perceived workload is low. In this paper, we describe a data acquisition system and machine learning based algorithms to determine perceived workload. Data collected were from on-road driving in light and heavy traffic, and individual physiological measures were recorded while the driver also performed in-vehicle tasks. Initial results show how the workload function can be personalized to an individual, and what implications this may have for vehicle design.

Lane change events can be a source of traffic accidents; drivers can make improper lane changes for many reasons. In this paper we present a comprehensive study of a passive method of predicting lane changes based on three physiological signals: electrocardiogram (ECG), respiration signals, and galvanic skin response (GSR). Specifically, we discuss methods for feature selection, feature reduction, classification, and post processing techniques for reliable lane change prediction. Data were recorded for on-road driving for several drivers. Results show that the average accuracy of a single driver test was approx. 70%. It was greater than the accuracy for each cross-driver test. Also, prediction for younger drivers was better.