SYSTEM AND METHOD FOR DETECTING PARAMETRIC OUTLIERS IN RELATION TO VEHICLE PULL/DRIFT DATA

A method of identifying parametric outliers associated with vehicle pull/drift compensation may include receiving fleet data from a plurality of vehicles, the fleet data including pull/drift steering data associated with the plurality of vehicles, processing the pull/drift steering data to generate...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Lysenko, Michael George, Omrani, Ala Eddine, Jammoussi, Hassene, Slavin, Daniel, Deshpande, Shreshta Rajakumar
Format: Patent
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:A method of identifying parametric outliers associated with vehicle pull/drift compensation may include receiving fleet data from a plurality of vehicles, the fleet data including pull/drift steering data associated with the plurality of vehicles, processing the pull/drift steering data to generate cleaned pull/drift steering data, and employing a machine learning module to define, for each of a plurality of respective different vehicle classifications, an outlier threshold for comparison to the cleaned pull/drift steering data to identify parametric outliers, and to identify the parametric outliers within each vehicle classification based on the comparison.