Systems and methods for traffic learning are described. One embodiment of a method includes predicting a vehicle route of a vehicle, determining historical traffic data related to at least one segment along the vehicle route, and determining an energy traffic budget for electric power consumption of the vehicle along the vehicle route, where the energy traffic budget is learned from the historical traffic data. Some embodiments include creating a schedule for transitioning between gasoline power and electric power for the vehicle route, implementing the schedule as the vehicle traverses the vehicle route, and in response to encountering traffic along the vehicle route, implementing the energy traffic budget.