Design and research of underwater multi-agent cooperative getting-to-the-firing-position decision-making algorithm based on MoZi platform

Abstract

ObjectivesThis study focuses on the multi-agent cooperative getting-to-the-firing-position decision-making algorithm design of underwater battlefields based on the MoZi platform, and explores the combat application of underwater unmanned platforms. MethodsA design idea based on the MoZi platform is proposed. For action decision-making under deterministic rules, the design of knowledge graph modules is discussed. The Neo4J database is used to build a global knowledge database, and a data exchange interface is designed to provide the auxiliary decision information input of deterministic rules for the intelligent algorithm module. For action decision-making under uncertain and incomplete information, the design of a cooperative space occupying a decision-making algorithm based on MADDPG is discussed under the set scenario, and the design concept of a state space, action space, reward function and interaction with the knowledge graph module is put forward. ResultsThe overall design of the algorithm framework is completed on the basis of the abovementioned ideas. The integration of the knowledge graph and intelligent algorithm is realized on the MoZi platform, and the effectiveness of the integration of behavior constraints under deterministic rules and the intelligent algorithm for processing uncertain situation information is further verified through the visualization provided by the platform.ConclusionsThis paper can provide valuable references for the design of multi-agent underwater getting-to-the-firing-position cooperative algorithms.

Keywords

• multi-agent
• mozi platform
• knowledge graph
• maddpg