In Vehicular Ad-hoc Networks (VANETs), increasing service demand complicates vehicle network control. Clustering management simplifies this control and enhances spectrum resource utilization. This paper introduces a distance difference prediction-based Vehicle-to-Vehicle (V2V) multicast cluster transmission mechanism using an interacting multiple model Kalman filter, termed DP-IMMK. It predicts distance differences between vehicles by considering three distinct movement statuses and clusters vehicles based on V2V transmission distance and multicast performance. Information is transmitted within clusters via V2V multicast, and cluster maintenance is managed by setting clustering factors and predicting vehicle movement statuses. Simulation results indicate a prediction error of less than 1.81%. Compared to geographical static clustering algorithms, this mechanism improves system energy efficiency by 23.61%.