Task Scheduling for Warship Formation Air Defense

As data capacity increasing and artificial intelligence technology improving, transforming data into knowledge to provide solutions for task scheduling has become a viable direction. Knowledge graph has been widely used in recommendation systems due to its comprehensibility of its tuple, but it is seldom used in task scheduling. In this paper, a task scheduling method based on knowledge graph and graph neural network is proposed and compare with Optimization Algorithm. This method uses knowledge graph to record historical tasks, and uses graph neural network to calculate the environmental characteristics of each task. The result shows that the cost time for each task scheduling process with this method is shorter and the adaptability to environmental changes is better than Optimization Algorithm, which means this method is more suitable for the requirements for complex battlefield situations.