On-board prediction of future speed profile for energy management of hybrid electric vehicles

Vehicular communications could be exploited for energy management of vehicles. We propose a system which provides that a vehicle estimates its future speed profile gathering status messages broadcasted by the surrounding vehicles and/or the infrastructure and inputting them in a traffic simulator used as a predictor. The system has been validated by simulation considering an urban scenario inspired to the Ecotekne campus at the University of Salento and a Manhattan scenario, very challenging in relation to the prediction of the speed profile. Simulation results have shown that the prediction error is quite low for the first scenario. In the Manhattan scenario, the error is quite high in case each vehicle limits itself to send messages only to its neighbours and does not transmit the information regarding its route. However, the error can be significantly reduced if route information is broadcasted and the infrastructure relays the messages transmitted by vehicles. The proposed system has been tested in the Ecotekne campus.