TY - GEN
T1 - Applying Bayesian nonparametrics to non-homogeneous driving operation data towards prediction
AU - Hamada, Ryunosuke
AU - Kubo, Takatomi
AU - Ikeda, Kazushi
AU - Zhang, Zujie
AU - Shibata, Tomohiro
AU - Bando, Takashi
AU - Egawa, Masumi
PY - 2013/1/1
Y1 - 2013/1/1
N2 - Prediction of driving behaviors is important problem in developing the next-generation driving support system. In order to take account of diverse driving situations, it is necessary to deal with multiple time series data considering commonalities and differences among them. In this study we utilize the beta process autoregressive hidden Markov model (BP-AR-HMM) that can model multiple time series considering common and different features among them using the beta process as a prior distribution. We apply BP-AR-HMM to actual driving operation data to estimate vector-autoregressive process parameters that represent the segmental driving behaviors, and with the estimated parameters we predict the driving behaviors of unknown test data. Prediction accuracy of test data using BP-AR-HMM is compared with that of using classical HMM. The results suggest that it is possible to identify the dynamical behaviors of driving operations using BP-AR-HMM, and with BP-AR-HMM we can predict driving behaviors better in actual environment than with HMM.
AB - Prediction of driving behaviors is important problem in developing the next-generation driving support system. In order to take account of diverse driving situations, it is necessary to deal with multiple time series data considering commonalities and differences among them. In this study we utilize the beta process autoregressive hidden Markov model (BP-AR-HMM) that can model multiple time series considering common and different features among them using the beta process as a prior distribution. We apply BP-AR-HMM to actual driving operation data to estimate vector-autoregressive process parameters that represent the segmental driving behaviors, and with the estimated parameters we predict the driving behaviors of unknown test data. Prediction accuracy of test data using BP-AR-HMM is compared with that of using classical HMM. The results suggest that it is possible to identify the dynamical behaviors of driving operations using BP-AR-HMM, and with BP-AR-HMM we can predict driving behaviors better in actual environment than with HMM.
KW - Bayesian nonparametric approach
KW - beta process
KW - beta process autoregressive hidden Markov model
KW - driving behavior prediction
UR - http://www.scopus.com/inward/record.url?scp=84901376532&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84901376532&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84901376532
SN - 9780992862602
T3 - European Signal Processing Conference
BT - 2013 Proceedings of the 21st European Signal Processing Conference, EUSIPCO 2013
PB - European Signal Processing Conference, EUSIPCO
T2 - 2013 21st European Signal Processing Conference, EUSIPCO 2013
Y2 - 9 September 2013 through 13 September 2013
ER -