Fatigue Measurement of Task: Based on Multiple Eye-Tracking Parameters and Task Performance

摘要

Human error can be reduced by monitoring fatigue during work and taking measures so that accidents can be effectively prevent. In this paper, participants accumulated fatigue during the process of using the experimental program integrating the open source SDK, and their task performance and PERLCLOS were collected and calculated by the experimental program. Finally, a neural network model was established to describe the relationship between various parameters and fatigue. Through experiments, it is found that PERCLOS and task performance can well reflect fatigue state in tasks of searching. This model can detect the fatigue degree of different people in specific application scenarios.