Deep Model for Improved Operator Function State Assessment
Document Type
Conference Paper
Publication Date
2014
Publication Title
Proceedings of the 8th Annual Modeling, Simulation & Visualization Student Capstone Conference
Pages
58-59
Conference Name
8th Annual Modeling, Simulation & Visualization Student Capstone Conference, April 17, 2014, Suffolk, Virginia
Abstract
A deep learning framework is presented for engagement assessment using EEG signals. Deep learning is a recently developed machine learning technique and has been applied to many applications. In this paper, we proposed a deep learning strategy for operator function state (OFS) assessment. Fifteen pilots participated in a flight simulation from Seattle to Chicago. During the four-hour simulation, EEG signals were recorded for each pilot. We labeled 20- minute data as engaged and disengaged to fine-tune the deep network and utilized the remaining vast amount of unlabeled data to initialize the network. The trained deep network was then used to assess if a pilot was engaged during the four-hour simulation.
Original Publication Citation
Li, F., Wen, J., Li, J., Zhang, G., Xu, R., & Schnell, T. (2014). Deep model for improved operator function state assessment. In Proceedings of the 8th Annual Modeling, Simulation & Visualization Student Capstone Conference (pp. 58-59), Old Dominion University. https://digitalcommons.odu.edu/cgi/viewcontent.cgi?article=1021&context=msvcapstone#page=59
Repository Citation
Li, Feng; Wen, Jonathan; Li, Jiang; Zhang, Guangfan; Xu, Roger; and Schnell, Tom, "Deep Model for Improved Operator Function State Assessment" (2014). Electrical & Computer Engineering Faculty Publications. 372.
https://digitalcommons.odu.edu/ece_fac_pubs/372
ORCID
0000-0003-0091-6986 (Li)
Comments
Link to proceedings as a whole: https://digitalcommons.odu.edu/msvcapstone/proceedings/2014/1