Unmanned Swarm System Situation Awareness Based on Dynamic Bayesian Network

In a multidimensional and complex battlefield environment, unmanned swarm system can determine the threat level of surrounding targets through situation awareness to assist in guiding strategic decision-making. This paper constructs a model of unmanned swarm system situation awareness based on Dynamic Bayesian Network. By analyzing the threat situation caused by the different behavior states of unmanned swarm system, the time-varying probability of different behavioral intentions is calculated to assist decision-making more effectively. The advantages of this model are comprehensively considering the changes in the time dimension and the probability characteristics of multiple state factors of unmanned swarm system. And the effectiveness of the model is verified by experimental simulation, which indicates the strong application value of the model.

I. Introduction

Unmanned swarm system is the key direction of the development of advanced military technology. Combining unmanned swarm system with land, sea and air combat platforms to achieve swarm combat capabilities can have an immeasurable potential impact on future military operations. Under complex battlefield conditions, unmanned swarm system needs to obtain real-time information to comprehensively perceive the battlefield situation and determine the threat level of surrounding targets. The main task of situation awareness is to use drones and other sensors to obtain information data, and match situation information with knowledge rules to make auxiliary decisions [1]. Situation awareness plays a guiding role as the basis for swarm decision-making, control and confrontation.