Power management for Plug-in Hybrid Electric Vehicles using Reinforcement Learning with trip information

In this paper, we present a new method of power management for Plug-in Hybrid Electric Vehicles (PHEVs) using Reinforcement Learning technique combined with trip information. Our new control strategy uses the remaining travel distance, which can be easily obtained from today's Global Positioning System (GPS), for the energy optimization of PHEVs. For a given trip, the remaining distance is highly correlated to the future energy consumption, a quantity our controller tries to learn and optimize continuously. The simulation results confirm the self-improving capability of our reinforcement learning controller and show that our controller outperforms the rule-based controller with respect to a defined reward function.