TY - GEN
T1 - Human computer interface using electroencephalography for driver behavior classification
AU - Manchala, Vamsi K.
AU - Clara, Alvaro V.
AU - Subramanian, Susheelkumar C.
AU - Redkar, Sangram
AU - Sugar, Thomas
N1 - Publisher Copyright:
Copyright © 2019 ASME.
PY - 2019
Y1 - 2019
N2 - It is important to know and be able to classify the drivers’ behavior as good, bad, keen or aggressive, which would aid in driver assist systems to avoid vehicle crashes. This research attempts to develop, test, and compare the performance of machine learning methods for classifying human driving behavior. It also proposes to correlate driver affective states with the driving behavior. The major contributions of this work are to classify the driver behavior using Electroencephalograph (EEG) while driving simulated vehicle and compare them with the behavior classified using vehicle parameters and affective states. The study involved both classical machine learning techniques such as k-Nearest Neighbor (KNN), Support Vector Machine (SVM), Artificial Neural Network (ANN) and latest “unsupervised” Hybrid Deep Learning techniques, and compared the accuracy of classification across subjects, various driving scenarios and affective states.
AB - It is important to know and be able to classify the drivers’ behavior as good, bad, keen or aggressive, which would aid in driver assist systems to avoid vehicle crashes. This research attempts to develop, test, and compare the performance of machine learning methods for classifying human driving behavior. It also proposes to correlate driver affective states with the driving behavior. The major contributions of this work are to classify the driver behavior using Electroencephalograph (EEG) while driving simulated vehicle and compare them with the behavior classified using vehicle parameters and affective states. The study involved both classical machine learning techniques such as k-Nearest Neighbor (KNN), Support Vector Machine (SVM), Artificial Neural Network (ANN) and latest “unsupervised” Hybrid Deep Learning techniques, and compared the accuracy of classification across subjects, various driving scenarios and affective states.
KW - Brain computer interface
KW - Driver behavior classification
UR - http://www.scopus.com/inward/record.url?scp=85076468361&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85076468361&partnerID=8YFLogxK
U2 - 10.1115/DETC2019-97540
DO - 10.1115/DETC2019-97540
M3 - Conference contribution
AN - SCOPUS:85076468361
T3 - Proceedings of the ASME Design Engineering Technical Conference
BT - 21st International Conference on Advanced Vehicle Technologies; 16th International Conference on Design Education
PB - American Society of Mechanical Engineers (ASME)
T2 - ASME 2019 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, IDETC-CIE 2019
Y2 - 18 August 2019 through 21 August 2019
ER -