Detecting Stress During Real-World Driving Tasks Using Physiological Sensors

Abstract

This paper presents methods for collecting and analyzing physiological data during real world driving tasks to determine a driver's relative stress level. Electrocardiogram, electromyogram, skin conductance and respiration were recorded continuously while drivers followed a set route through open roads in the greater Boston area. Data from twenty-four drives of at least fifty minute duration were collected f or analysis. In Analysis I features from five minute intervals of data were used to distinguish three levels of driver stress with an accuracy of over 97% across multiple drivers and driving days. In Analysis II, continuous physiological features were correlated with a continuous metric of observable stressors showing that on a real-time basis metrics of skin conductivity and heart rate were most closely correlated with driver stress level. Such automatically calculated physiological features could be used to help manage non-critical in-vehicle information systems and improve the driving experience.

Rorschach Performance Assessment System (R-PAS) and vulnerability to stress: A preliminary study on electrodermal activity during stress

“…The increment of the average EDA during recovery over the average EDA during baseline was labeled "EDA Change: Recovery minus Baseline," and was used as proxy marker for prolonged, sympathetic arousal following stressful stimulation. This procedure follows previous EDA studies with baseline, stress, and/or recovery conditions (Healey and Picard, 2005;El-Sheikh et al 2007;Reinhardt et al, 2012).…”

Activity-Aware Mental Stress Detection Using Physiological Sensors

“…For instance, Healey and Picard measured drivers' stress reactions by monitoring multiple physiological signals, such as ECG, GSR, electromyogram (EMG), and respiration in a prescheduled route setting [10]. They used 5-minute intervals of data during the rest, highway, and city driving conditions to distinguish between three levels of driver stress.…”

Assessment of Driver Stress from Physiological Signals collected under Real-Time Semi-Urban Driving Scenarios

“…Physiological features have been extracted from signals like electrocardiogram (ECG / EKG), electromyogram (EMG), galvanic skin response (GSR), respiration and electrocardiogram (EKG / ECG) and classified for detecting stress level or emotions of automotive drivers [5,6,7] . The mostly used physiological parameters have been (a) heart rate (HR) and heart rate variability (HRV), both derived from ECG and (b) skin conductance responses (SCR) and other related features which account for sudden stress responses, derived from GSR.…”

“…The salient features of our stress level detection approach includes: (a) first work to the best of our knowledge focusing a developing country like India in terms of scenario design, subject population and road settings, (b) physiological data collected in realtime driving scenarios modeled the stress contributing factors into a multiclass problem instead of a binary class 9 , (c) an exhaustive set of physiological features (39 statistical, syntactic and spectral) were extracted representing driver's current physiological state, (d) instead of less number of subjects and a one-fold classification [6][7] , we used more subjects and performed analysis on single as well multi-turn drive data, and (e) six neural network 4-class classifiers were evaluated instead of generalizing a single classifier.…”