Personalized driving assistance algorithms: Case study of federated learning based forward collision warning

•Proposed a federated learning framework to develop personalized driving assistance models.•Trained forward collision warning (FCW) models through collecting updated model parameters rather than raw data.•Introduced a driver-specific batch normalization (BN) layer to allow models adapting to heterog...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Accident analysis and prevention 2022-04, Vol.168, p.106609-106609, Article 106609
Hauptverfasser:	Yu, Rongjie, Zhang, Ruici, Ai, Haoan, Wang, Liqiang, Zou, Zihang
Format:	Artikel
Sprache:	eng
Schlagworte:	Accidents, Traffic - prevention & control Algorithms Automobile Driving Batch normalization Federated learning Forward collision warning Humans Long short-term memory Machine Learning Personalized driving assistance algorithms Protective Devices Reproducibility of Results
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!