Knowledge of the upcoming traffic velocity along a route can help in many respects, among them optimizing energy management for hybrid vehicles, which, for instance, could reduce instantaneous battery usage if a traffic jam is upcoming in the next future. While such kind of knowledge can hardly be precise on a single-vehicle level, we show in this paper that a prediction method which combines present and past Vehicle-to-Everything (V2X) information can strongly improve the energy efficiency. Our approach is first compared with other prevailing prediction methods and its advantages in terms of stability and accuracy are shown. Then the prediction results are applied in a hybrid powertrain control example, in which its potential in fuel savings are illustrated.
V2X Database Driven Traffic Speed Prediction
19.09.2021
1784109 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Traffic speed prediction device and traffic speed prediction method
Europäisches Patentamt | 2025
|Traffic Speed Prediction with Neural Networks
British Library Conference Proceedings | 2017
|Traffic speed prediction method based on multi-moment traffic flow
Europäisches Patentamt | 2025
|Traffic speed prediction using deep learning method
IEEE | 2016
|