A method for scheduling EV battery swap based on IoV, in which the roadside units regard battery-swap station clusters with a high degree of potential cooperation as a whole, and gather them into a single battery swap area; taking a service rate of a cluster of swap stations as an assessment target, and mainly examining a service capability, a service quality, and location information of each of swap station nodes themselves, and a current state of those electric vehicles that require to swap battery; providing a solution of the best joint actions for the overall electric vehicles to maintain the overall service balance of every swap station and improve a long-term performance of Internet of Vehicles. According to the invention, a battery of the electric vehicles can be swapped as soon as possible, and every battery swap station can maintain business balance.