Abstract: The present disclosure provides methods, systems, and devices for controlling electric vehicle charging across multiple customers and multiple fleets of electric vehicles. These methods, systems, and devices may implement machine learning to determine distinct charging strategies for a plurality of charging depots. Scheduling methods systems, and devices disclosed herein do not require fixed electric vehicle arrival times but may instead update charging strategies in real time based on changes in a state of a system.

Abstract: A method of controlling battery preconditioning in a vehicle includes determining a first preconditioning characteristic of a first battery of a first vehicle relative to a charging station. The method further includes transmitting the first preconditioning characteristic. The method further includes receiving, from a second vehicle, a second preconditioning characteristic of a second battery relative to the charging station. The method further includes comparing the first preconditioning characteristic and the second preconditioning characteristic to determine a preconditioning ranking for the first vehicle and the second vehicle. The method further includes determining a queue of the first and second vehicles for the charging station using the preconditioning ranking of the first and second vehicles.