Research On Multi-agent Cooperative Target Search And Coverage Strategy

Due to its strong self-organization,flexibility,and scalability,multi agent systems are widely used in regional monitoring,unmanned search and rescue,and target tracking scenarios,Especially the multi agent control system based on unmanned aerial vehicles.These application scenarios require multi agent systems to have efficient and fast multi target search coverage capabilities.On this basis,multi agent nodes are also required to have flexible mobility,efficient task processing capabilities,and the ability to adapt to constantly changing surrounding environments.However,in practical application scenarios,there are often many hard constraints and uncertainties.Under the influence of these constraints and uncertainties,such as uncertain distribution of environmental targets,low total energy,communication connectivity and maintenance of agent systems,and complex node operating environments,the progress of multi agent systems in completing tasks will be hindered.With the continuous development of technology,the requirements of society for various capabilities of multi agent systems are gradually increasing.How to design a multi agent cooperative target search coverage strategy suitable for practical applications to efficiently perform tasks is of great significance.Therefore,this paper selects three multi-agent task scenarios,and designs multiagent cooperative target search coverage strategies for different tasks based on control theories and optimization algorithms such as graph theory,harmony search,and convex optimization,respectively,for three different hardware conditions(unmanned aerial vehicles)and environmental conditions.The specific research content is as follows:In Task 1,this paper studies the multi agent target search problem in areas with uncertain target distribution.This paper designs a method of integrating environmental information and agent perception information into search graphs,using state consistency to quickly converge the search graphs maintained by each agent node.Based on the above conditions,an optimization function is designed that considers local search returns and future search returns.Using DMPC method to make motion decisions for each agent,a multi agent cooperative target search strategy based on fixed wing UAV is proposed.In Task 2,this paper studies the multi agent target coverage problem based on rotor UAV with connectivity preserving constraints.In this paper,we first model and analyze the problem,and then design an agent benefit function based on global information.Each agent can choose a locally optimal connection keeping deployment decision based on this benefit function.Based on this,a multi-agent target coverage strategy based on rotor UAV is proposed by using an improved harmony search algorithm to traverse the global optimal agent network location.In Task 3.this paper studies the multi agent target coverage problem based on fixed wing unmanned aerial vehicles with optimal energy consumption.This paper first analyzes the problem by modeling,and then designs a multi agent multi target energy saving allocation mechanism.On this basis,it analyzes the multi agent target coverage trajectory problem based on fixed wing unmanned aerial vehicles,proposes two coverage trajectory schemes,and utilizes the successive convex optimization(SCA)method to optimize the energy saving of the proposed two coverage trajectories.This paper simulates and analyzes the multi agent cooperative target search and coverage strategies proposed in the above three scenarios.The results show that the strategy proposed in this paper can efficiently and satisfy constraints to complete specified tasks.