Optimization of Two-Phase Sampling Designs with Application to Naturalistic Driving Studies
Journal article, 2021

Naturalistic driving studies (NDS) generate tremendous amounts of traffic data and constitute an important component of modern traffic safety research. However, analysis of the entire NDS database is rarely feasible, as it often requires expensive and time-consuming annotations of video sequences. We describe how automatic measurements, readily available in an NDS database, may be utilised for selection of time segments for annotation that are most informative with regards to detection of potential associations between driving behaviour and a consecutive safety critical event. The methodology is illustrated and evaluated on data from a large naturalistic driving study, showing that the use of optimised instance selection may reduce the number of segments that need to be annotated by as much as 50%, compared to simple random sampling.

optimal design

unequal probability sampling

pseudo-likelihood

safety critical event

naturalistic driving studies

case-control studies

Author

Henrik Imberg

Chalmers, Mathematical Sciences, Applied Mathematics and Statistics

Vera Lisovskaja

Chalmers, Mathematical Sciences, Applied Mathematics and Statistics

Selpi Selpi

Chalmers, Mechanics and Maritime Sciences, Vehicle Safety, Crash Analysis and Prevention

Olle Nerman

Chalmers, Mathematical Sciences, Applied Mathematics and Statistics

IEEE Transactions on Intelligent Transportation Systems

1524-9050 (ISSN)

Vol. In Press 1-14

Statistical methods to assess driving behaviour and causation of accidents from large naturalistic driving studies

Swedish Research Council (VR), 2012-01-01 -- 2015-12-31.

Areas of Advance

Transport

Subject Categories

Probability Theory and Statistics

DOI

10.1109/TITS.2020.3038180

More information

Latest update

2/25/2021