Research On Task Assignment Technology Of Air-to-Air Missile

In modern warfare,air dominance has important strategic significance,and air-to-air missiles are important weapons in the fight.In an air combat,there are often multiple targets with strong maneuverability.It is difficult to adapt to this scenario for the combat mode of a single missile attacking a single target,while cluster operations of air-to-air missiles is an important direction of air combat.In such mode,the task assignment of air-to-air missiles has important practical significance for improving the combat performance.This article introduces and analyzes the characteristics and historical development of air-to-air missiles,and provides theoretical and factual basis for the establishment of air-to-air missiles task assignment problem.This article introduces the background knowledge of air-to-air missiles,including the guidance system of missiles and the main problems air-to-air missiles faced in actual combat.For the air-to-air missile task assignment,this paper considers the characteristics of each stage of cluster operations of air-to-air missiles.Considering the air combat relationship between carrier and target,missile and target,and considering the operational constraints in the air combat environment,the combat phase is divided into three phases: a pre-attack phase in which the aircraft flies to the attackable area,an attack phase in which the air-to-air missile attacks,and a detection and damage evaluation phase before and after the air-to-air missile launch.This paper established the mixed integer programming model,the assignment model and the traveling salesman problem model.Finally,this paper researches related algorithms for solving above models.The main research contributions of this paper are as follows:1.For the task assignment problem in the pre-attack stage,a hybrid integer programming model was established with three optimization goals: maximum situational advantage,shortest time to enter attackable area,and similar time to enter attackable area.Based on the existing papers and research of the characteristics of this problem,this paper designs some basic heuristic operators,and designs a variety of heuristic algorithms under a rule-based heuristic algorithm framework,a meta-heuristic algorithm framework,and a hyper-heuristic algorithm framework.Finally,this paper compares and analyzes the accuracy and convergence of above algorithms.For the problem of dynamic battlefield environment,a method of dynamic heuristic algorithm design based on the same population was proposed.This dynamic heuristic algorithm design method shares the same population in the existing heuristic algorithms,which has good practicability and scalability.2.For the assignment model of the air-to-air missile in the attack phase,solutions based on bipartite graph matching and market auction theory are studied.An algorithm based on linear programming theory and roulette strategy is proposed,with a partial proof of optimality and convergence.Finally,it is applied to the task assignment of air-to-air missiles,which has a good real-time performance.3.For the detection and damage assessment problems before and after the attack,a traveling salesman problem model is established,and a highly accurate iterative local search algorithm is designed based on the heuristic rule operator.For large-scale problems,the performance of heuristic algorithms degrades.An improved self-organizing mapping neural network algorithm with a hybrid strategy of Winner Takes Most(WTM)and Winner Takes All(WTA)mechanisms is proposed.In this method,the distance between the neuron nodes is represented by the difference between the serial numbers.A Gaussian kernel function is used to describe the influence between the neuron nodes.Through the iterative update of the neuron position,the mapping relationship between the neuron and the detection point is established,and a solution of the problem is obtained.