TY - GEN
T1 - Building representative velocity profiles using FastDTW and spectral clustering
AU - Lohrer, Jürgen
AU - Lienkamp, Markus
N1 - Publisher Copyright:
© 2015 IEEE.
PY - 2016/1/8
Y1 - 2016/1/8
N2 - The use of map based representative velocity profiles allows to predict the future state of a vehicle. The suggested approach is based on Fast Dynamic Time Warping. Spectral clustering is used to distinguish velocity profiles. Applying abstraction can significantly reduce computation time with a minor effect on cluster allocation. Outlier removal increases the quality of cluster identification. The approach was applied to the road network of Munich, to prove the universal applicability.
AB - The use of map based representative velocity profiles allows to predict the future state of a vehicle. The suggested approach is based on Fast Dynamic Time Warping. Spectral clustering is used to distinguish velocity profiles. Applying abstraction can significantly reduce computation time with a minor effect on cluster allocation. Outlier removal increases the quality of cluster identification. The approach was applied to the road network of Munich, to prove the universal applicability.
KW - Fast Dynamic Time Warping
KW - Spectral Clustering
KW - Speed Profiles
KW - Time Series Classification
UR - http://www.scopus.com/inward/record.url?scp=84966545998&partnerID=8YFLogxK
U2 - 10.1109/ITST.2015.7377398
DO - 10.1109/ITST.2015.7377398
M3 - Conference contribution
AN - SCOPUS:84966545998
T3 - 2015 14th International Conference on ITS Telecommunications, ITST 2015
SP - 45
EP - 49
BT - 2015 14th International Conference on ITS Telecommunications, ITST 2015
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 14th International Conference on ITS Telecommunications, ITST 2015
Y2 - 2 December 2015 through 4 December 2015
ER -