TY - GEN
T1 - Driver Behavior Classification in Crash and Near-Crash Events Using 100-CAR Naturalistic Data Set
AU - Abdelrahman, Abdalla
AU - Abu-Ali, Najah
AU - Hassanein, Hossam S.
N1 - Funding Information:
ACKNOWLEDGEMENTS This work is funded by project no. 31R014-Research Center RTTSRC-4-2013 provided by the Roadway Transportation & Traffic Safety Research Center, United Arab Emirates University. This research is also supported by a grant from the Natural Sciences and Engineering Research Council of Canada (NSERC) under grant number: STPGP 479248.
Publisher Copyright:
© 2017 IEEE.
PY - 2017/7/1
Y1 - 2017/7/1
N2 - Recently, several car insurance companies got interested in classifying the behavior of drivers. Usage-based insurance (UBI), such as Pay-How-you- Drive (PHYD) scheme, is an innovative idea in which the insurance premium changes based on the driving behavior. This behavior is usually evaluated in terms of vehicle-related variables such as distance, speed, and acceleration to determine the expected risk profile for drivers. In this paper, an additional level of classification in the hierarchy of profiling is proposed. Using the 100-CAR naturalistic driving study (NDS) data set, five different Hidden Markov Models (HMMs) are trained to determine the fault responsibility of a Subject Vehicle (SV) in a crash or near-crash events. Two specific driving situations, which are conflicts with leading and following vehicles, are investigated in this study. Results show that these models can achieve a reasonable classification accuracy.
AB - Recently, several car insurance companies got interested in classifying the behavior of drivers. Usage-based insurance (UBI), such as Pay-How-you- Drive (PHYD) scheme, is an innovative idea in which the insurance premium changes based on the driving behavior. This behavior is usually evaluated in terms of vehicle-related variables such as distance, speed, and acceleration to determine the expected risk profile for drivers. In this paper, an additional level of classification in the hierarchy of profiling is proposed. Using the 100-CAR naturalistic driving study (NDS) data set, five different Hidden Markov Models (HMMs) are trained to determine the fault responsibility of a Subject Vehicle (SV) in a crash or near-crash events. Two specific driving situations, which are conflicts with leading and following vehicles, are investigated in this study. Results show that these models can achieve a reasonable classification accuracy.
UR - http://www.scopus.com/inward/record.url?scp=85046492834&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85046492834&partnerID=8YFLogxK
U2 - 10.1109/GLOCOM.2017.8253921
DO - 10.1109/GLOCOM.2017.8253921
M3 - Conference contribution
AN - SCOPUS:85046492834
T3 - 2017 IEEE Global Communications Conference, GLOBECOM 2017 - Proceedings
SP - 1
EP - 6
BT - 2017 IEEE Global Communications Conference, GLOBECOM 2017 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2017 IEEE Global Communications Conference, GLOBECOM 2017
Y2 - 4 December 2017 through 8 December 2017
ER -