Data-driven energy-efficient speed planning and adaptive car-following control for commuter plug-in hybrid electric vehicles

Abstract

This paper proposes a hierarchical adaptive eco-driving control scheme for commuter plug-in hybrid electric vehicles (PHEVs), which contains the data-driven speed planner in the upper layer and the speed tracking controller in the lower layer. Considering that the speed planning for PHEVs with the goals of fuel economy and comfort involving both coupled motion and powertrain dynamics can result in an enormous computational burden, a data-driven speed planner is designed based on traffic information of the commuter PHEV, where the energy consumption and battery state are described through two models by the neural networks, respectively. As such, the suggested speed ensuring fuel economy and comfort for the vehicle driving on this route can be obtained with an acceptable computational burden. Then, facing the immeasurable acceleration of the preceding vehicle and slow time-varying road grade, an adaptive car-following speed controller is designed through the backstepping recursive technique to track the suggested speed accurately and safely. The control effects can be adjusted through vehicle-to-cloud (V2C) communication. The simulation results demonstrate the effectiveness and advantages of the proposed scheme in improving fuel economy.