Research On Swarm Intelligence Task Allocation Method For UAV Swarms Combat System

Compared with manned aircraft,UAV has many advantages,such as light weight,low cost,no casualties,etc.It has outstanding advantages in the future battlefield.In the battlefield with complex environment,many strategic actions require the cooperation of multiple UAVs to execute.With the development of drone technology,the combat mode has gradually changed from single drone to drone cluster operations.Among them,task allocation technology is an important technology in UAV cluster operations,and it is the key to whether UAVs can obtain higher income on the basis of completing tasks.Therefore,UAV task assignment is a worthy research direction of UAV swarm operations.The problem of UAV task assignment is based on specific constraints,with the goal of achieving the optimal task efficiency of the UAV cluster,and assigning specific goals and tasks to each UAV.This article focuses on UAV task allocation technology,with the goal of achieving the optimal mission efficiency of the drone cluster,assign specific tasks to each UAV.Aiming at the task allocation technology of UAVs,this paper establishes task allocation models and designs effective swarm intelligence algorithms to solve the task allocation problem of multiple UAVs,taking into account the UAV performance constraints,so as to obtain the task assignment plan that makes the UAV cluster combat efficiency the highest.The main content of this article can be summarized as follows:(1)In view of the shortcomings in the existing multi-UAV task allocation model,we take into account the constraints of the resources carried by the drone and the number of tasks that can be completed.Taking the mission effectiveness of the three tasks of reconnaissance,attack and damage evaluation as the objective function,we propose a new UAV task assignment model.In order to efficiently solve the optimal task allocation scheme,we design a quantum pigeon-inspired optimation algorithm.Based on the pigeon-inspired optimation algorithm,we introduce the quantum evolution mechanism to solve the shortcomings of the pigeon swarm algorithm,such as insufficient development and exploration capabilities and easy to fall into local convergence.And compared with the traditional swarm intelligence algorithms,we can verify the validity of the designed algorithm.(2)We study the decision-making problem when there is conflict between the decisions of the two parties in the UAV cluster operation through using game theory to analyze operational decisions.We propose a UAV combat decision-making method based on game theory,it transforms the UAV combat decision-making problem into the Nash equilibrium solution of the mixed strategy in the process of solving the game.In order to solve the problem efficiently,we combined quantum computing theory and swarm intelligence optimization algorithm to design the quantum krill herd algorithm.Through a number of commonly used benchmark functions,we verify the efficiency of the quantum krill swarm algorithm.Finally,we use the quantum krill herd algorithm to solve the game decision-making problem of UAV cluster operations.Simulation experiments show that the proposed quantum krill herd algorithm can effectively solve the Nash equilibrium solution in the UAV game between the two combatants.(3)We study the UAV task allocation problem when there are multiple targets that need to be optimized at the same time.Aiming at the shortcoming that the single-target UAV task allocation method can only optimize one target and the combined target under a certain set of weights and it can only give a task allocation plan,we designed a multi-target UAV task allocation model.Under the constraints of UAV performance,two goals are considered to optimize at the same time,namely,minimizing the total value of the undestroyed UAV target and the loss cost of UAV.We design a multi-objective quantum krill herd algorithm to solve this multi-objective UAV task assignment problem.Finally,we can get the Pareto solution set including different task allocation plans.Decision makers can choose the appropriate task allocation plan according to the importance of the target in practice.