Detecting stress during real – world driving tasks based on physiological signals using time-frequency distribution

Ravindar, Jeyaahthaarsine (2021) Detecting stress during real – world driving tasks based on physiological signals using time-frequency distribution. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

	Text (24 Pages) Detecting stress during real – world driving tasks based on physiological signals using time-frequency distribution.pdf - Submitted Version Download (930kB)
	Text (Full Text) Detecting stress during real – world driving tasks based on physiological signals using time-frequency distribution.pdf - Submitted Version Restricted to Repository staff only Download (3MB)

Abstract

Driving stress can be defined as any kind of change which causes physical, emotional, or psychological strain during driving. Stress is the body's response of a person to anything that needs attention or action. A driver’s excess stress can affect the driving performance and causes an increment in crash likelihood. The level of stress of a driver can be vary depending on the different conditions of driving. Physiologically, stress directly can be measured by the changes occur in the skin conductance, heart rate, respiration and muscle activities. The main goal of this study is to understand driving stress during real-world driving tasks based on the physiological signals at different driving locations using time-frequency distribution (TFD). In this project, there are 9 subjects (driver records) of at least 60 minutes duration. The physiological signals involved in this project are the electrocardiogram (ECG), electromyogram (EMG), foot galvanic skin response (Foot GSR). Three stages are implemented in this analysis of the project which are known as signal pre-processing, signal processing and signal classification. In signal pre-processing state, filtering the signals using a bandpass filter is important to decrease and also to smooth out high-frequency noise that is associated with measurements. Hence, the pre-processing step is implemented to reduce the noise impacts which can affect the interpretation of the signals. Next, in TFD, adjusting the time and frequency resolution is a must so that the important information of the physiological signals can be obtained. The average

Actions (login required)

View Item