Drivers' workload classification through electrocardiography

• Main Author: Jiang, Xinlai
• Other Authors: Huang Guangbin
• Format: Final Year Project
• Language: English
• Published: 2016
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/68124
• Institution: Nanyang Technological University

Currently, many new technologies are added to the vehicle system and provide user-friendly functions. More and more people are distracted by these new functions, and as a result the number of accidents increases. Therefore, Understanding drivers’ workload is important for evaluating these functions. This report demonstrates the classification of human workload through detecting and analyzing Electrocardiography (ECG) signals of humans when they are driving. This project has two stages: the first stage is to conduct experiments by using a driving simulator; the second stage is to interpret the ECG data and classify the cognitive load through ECG by the machine learning algorithm called Extreme Learning Machines. Finally, the results show that the classification accuracy is of up to 93% for detecting levels of workload. Compared to the subjective measurement of workload, workload assessment via ECG is more accurate. This fact means that the classification process used here can assess the cognitive load by measuring ECG and in future can be embedded in the automobile system.