Research On Target Dynamic Hunting Method For Multi-USV

Multi-USV collaborative technology is an important research direction in the field of unmanned surface vehicles.And the Multi-USV hunting of targets is a hot topic in the field of multi-USV collaboration.This paper focuses on this issue.In this paper,firstly,a comprehensive description of the Multi-USV hunting problem is made,and the process of hunting is subdivided into two parts:the task assignment phase and the cooperation hunting phase.And then introduce the theories of biological immune system.Based on the immune characteristics manifested in the process of hunting,a layered Multi-USV architecture based on immune theory and Hybrid individual USV architecture is proposed.This structure enables the collaboration between multiple USVs and the collision avoidance of a single USV at the same time,improving unmanned combat capability and ability to avoid danger.Secondly,aiming at the task allocation during the process of hunting,the research on the method of task allocation is carried out.Transform the assignment problem into the choice of tasks for individual USV.The immune model of USV and mission based on immune theory is proposed.This model compares USV to B cell,compares the USV’s selection behavior to the task,compares the task to the antigen,The formal description of the target set translates the task selection problem into the calculation of the antibody concentration.Ultimately based on the size of the antibody concentration,the individual unmanned boats to make the task of choice.Then,based on the unique immune network theory,the immune network model in the collaborative round trapping process is constructed,and a collaborative hunting method based on this model is proposed.The method makes the mathematical description of antigen,encodes antibody.With the analysis of the hunting process,the antigen and antibody affinity,stimulating factor and restrain factor between antibodies calculation formula are constructed.using these factors to calculate the antibody concentration which determines USV’s behavior.Finally,the feasibility and superiority of the algorithms mentioned in this paper is verified through the design of simulation examples.