摘要：AbstractIt is advocated that the success and user acceptability of Fully Electric Vehicles (FEVs) will predominantly depend on their electrical energy consumption rate and the corresponding degree of autonomy that they offer. FEVs must provide their drivers with the highest possible autonomy, as well as with a high degree of reliability and robustness in terms of energy performance. Thus, appropriate innovative ICT solutions must be adopted, in order to assist the driver in dealing with such energy-related issues, and strengthen FEVs’ autonomy and reliability. Such an ICT solution tailored for FEVs is the focus of this paper. In detail, this paper presents a novel implementation of energy-efficient routing based on machine learning engines. It identifies and explains appropriate instance and target attributes, related to road segments and vehicles characteristics. Consequently, it proposes a robust machine learning model, capable of predicting the actual energy that is expected to be consumed by the vehicle on the particular road segment. Scalability aspects are, then, discussed and presented. Finally, useful conclusions on the performance of the proposed model are reached.
关键词：Fully Electric Vehicles;route calculation;energy efficiency;machine learing;neural networks
