The application of cyber-physical system (CPS) and edge computing in smart logistics systems has greatly improved the work efficiency. This article mainly focuses on the planning and scheduling problems in the CPS-based multiple automatic guided vehicles logistics system. A virtual network map with uneven time constraints is first proposed, which realizes the control and interaction of actual work in the control layer. Then, an improved A* path planning algorithm based on the characteristics of the path network model is implemented. A novel traffic control method is added to the planning process to achieve time-sensitive and proactive scheduling and collision avoidance. Results of the numerical experiments reveal that the proposed CPS-based heuristic method could be effectively applied to the multiple automatic guided vehicles logistics system to further improve system security as well as work efficiency.