Predictive modeling for vehicle time-series data

J. Wallner, F. Diermeyer, S. Engel

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The majority of data in current vehicles is evaluated solely in a local manner. The proposed method offers the possibility of using the latest data analysis techniques to develop functions for active safety and driver assistance systems, or vehicle testing of automated driving systems. We show an approach that combines the advantages of machine learning for logged vehicle data with the ability to use the predictive models created online in vehicles. Starting with intelligent data pre-processing, optimal conditions for the analysis step are established. Using machine learning techniques, predictive models are created to estimate various kinds of outcome variables. The concept is shown using the example of estimating the dynamic criticality in vehicles, based on driving dynamics signals and radar data.

Original languageEnglish
Title of host publicationAdvanced Vehicle Control AVEC’16 - Proceedings of the 13th International Symposium on Advanced Vehicle Control AVEC’16
EditorsJohannes Edelmann, Manfred Plochl, Peter E. Pfeffer
PublisherCRC Press/Balkema
Pages477-482
Number of pages6
ISBN (Print)9781315265285
StatePublished - 2017
Event13th International Symposium on Advanced Vehicle Control, AVEC 2016 - Munich, Germany
Duration: 13 Sep 201616 Sep 2016

Publication series

NameAdvanced Vehicle Control AVEC’16: Proceedings of the 13th International Symposium on Advanced Vehicle Control (AVEC'16)

Conference

Conference13th International Symposium on Advanced Vehicle Control, AVEC 2016
Country/TerritoryGermany
CityMunich
Period13/09/1616/09/16

Fingerprint

Dive into the research topics of 'Predictive modeling for vehicle time-series data'. Together they form a unique fingerprint.

Cite this