Track planning method, system and device based on V2V data sharing and medium

The invention discloses a trajectory planning method, system and device based on V2V data sharing and a medium. The method comprises the steps that map data is acquired, and sensing result data and positioning data of a vehicle and other vehicles are collected; establishing a self-vehicle coordinate system, and performing data frame timestamp matching, vehicle position compensation, vehicle position coordinate conversion and target matching de-duplication; a graph neural network is adopted to encode map information and historical track information of a vehicle and other vehicles, and a fully connected graph is constructed to realize information interaction; on the basis of a target driving track prediction model, in combination with vehicle-mounted unit communication collection information, prediction of future tracks of surrounding vehicles is completed, and scene vehicle time sequence position information is output; and the combined prediction and planning module outputs time sequence position information o