Evaluating the Measurement of Driver Heart and Breathing Rates from a Sensor-Equipped Steering Wheel using Spectrotemporal Signal Processing

Driver's status and behaviours such as inattention, drunk driving, or sleeping while driving play important roles in approximately half of all automobile crashes. For this reason, the last decade has seen an emergence of non-intrusive driver status monitoring systems with the ultimate goal of reducing the number of such accidents. From the different number of proposed methods, the use of the physiological signals, specifically the electrocardiogram (ECG), has shown useful. The acquisition of ECG signals during driving, however, presents a challenge due to movement artifacts, such as car and driver motion, and a good contact of the sensing electrodes, e.g., embedded on the driver seat. In this paper, we evaluate the ECG signals acquired from electrodes placed on the steering wheel under three aspects: (i) quality of the acquired signals; (ii) their usability to estimate an average and an instantaneous heart rate, and (iii) their usability to estimate the driver's breathing rate via innovative spectrotemporal processing of the acquired signals. Experimental results show that ECG signals obtained from the steering wheel have quality inline with that obtained from a benchmark chest ECG device, allow for both average and instantaneous heart rate to be measured, as well as breathing rate to be extracted.