Efficient Anticipatory Longitudinal Control of Electric Vehicles through Machine Learning-Based Prediction of Vehicle Speeds

Driving style and external factors such as traffic density have a significant influence on the vehicle energy demand especially in city driving. A longitudinal control approach for intelligent, connected vehicles urban areas is proposed this article to improve efficiency of automated The incorporates information from Vehicle-2-Everything communication anticipate behavior leading adapt accordingly. supervised learning derived train neural prediction model based recurrent network speed trajectories ego vehicles. For development, analysis evaluation approach, co-simulation environment presented that combines generic with microscopic simulation. This allows simulation different powertrains complex environment. investigation shows using V2X improves speeds significantly. can make use achieve more anticipatory driving which reduce consumption compared conventional Adaptive Cruise Control approach.