Analysis of Physiological Signals for Stress Recognition with Different Car Handling Setups

When designing a car, the vehicle dynamics and handling are important aspects, as they can satisfy purpose in professional racing, well contributing to driving pleasure safety, real perceived, regular drivers. In this paper, we focus on assessment of emotional response drivers while track with different car setups. The experiments were performed using dynamic simulator prearranged We recorded various physiological signals, allowing us analyze which setup is more influential terms stress arising subjects. logged two skin potential responses (SPRs), electrocardiogram (ECG) signal, eye tracking information. experiments, three setups used (neutral, understeering, oversteering). To evaluate how these affect drivers, analyzed their signals statistical tests (t-test Wilcoxon test) machine learning (ML) algorithms. results test show that SPR provide higher significance when evaluating among compared ECG signals. As for ML classifiers, count number positive or “stress” labels 15 s time intervals each subject particular setup. With support vector classifier, mean value four subjects equal 13.13% base setup, 44.16% oversteering 39.60% understeering end, our findings appears be least stressful, system enables effectively recognize configurations.