Research On Technologies Of Cooperative Attack And Defense Confrontation For Multi-uninhabited Smart Aerial Weapons

Uninhabited smart aerial weapon(USAW)will become the main tactical equipment of all regions,in the future warfare.Compared with traditional aerial weapons,USAW's most typical advantages lie in unmanned and intelligent.USAW can independently cooperate with each other according to unified command instructions or general operational rules,and implement coordinated and systematic combat forms in future offensive and defensive confrontations.Multi-USAW cooperative confrontation technologies are the main means to achieve the combat mode and improve operational effectiveness.This paper takes the basic tasks of reconnaissance,penetration and strike in actual combat as the background,and conducts in-depth research on the specific technical points under the coordinated attack and defense confrontation,and designs numerical simulation and experimental verification.The main work and innovations of this paper are summarized as follows:This paper studies the related technologies of cooperative mission planning and decision making for multi-USAW.For the cooperative reconnaissance mission planning problem of heterogeneous targets,consider the sensor model,heterogeneous target model and reconnaissance mission model of heterogeneous weapon platforms.The related constraints and performance indicators of the task assignment are given and the original cooperative investigation problem is described as a new multi-Dubins travelling saleman problem with task time window.The adaptive scaling factor is used to improve the original multi-objective symbiotic organisms search algorithm,which is used to optimize decision variables and generate optimal execution scenarios.Aiming at the problem of cooperative target allocation in air combat,this paper introduces the concept of dynamic attackable domain and non-escape domain under confrontation,studies its reference value in combat and introduces a fast fitting method based on BP neural network.This paper introduces a situational modeling method based on the dynamic attackable domain and the non-escape domain under the confrontation,which is used to uniformly describe the strike capability of the weapon platform for different types and different state targets within a certain region.Finally,based on the obtained attack situation,this paper describes the dynamic multi-weapon platform cooperative target allocation process as a "semi-dynamic" process with relatively unchanged situation,and proposes a centralized-distributed allocation decision method based on "Display Board" architecture.To achieve multiple rounds of attack distribution on a group of targets.This paper studies the related technologies of cooperative path planning in complex battlefield environment,analyzes various complex battlefield elements in the real environment,and carries out mathematical modeling and threat degree calculation for some environmental factors.Based on the terrain following/ terrain avoidance method,the single two-step track planning method is used in the complex battlefield environment.Based on this,a multi-weapon flight path planning algorithm in static environment is proposed to solve the problem for multi-weapon platforms under constant conditions.Finally,with reference to the traditional artificial potential field method,a method for constructing the uniformity and inferior potential field of the battlefield environment is designed,and based on this,a unified potential field is proposed.The multi-weapon platform cooperative path planning method is used to realize distributed online track planning of each weapon platform in a dynamic environment,so that the weapon platform can select the least disadvantaged and least threatous planning strategy according to the change of the battlefield environment in real time.This paper studies the related technologies of guidance and control in cooperative attack and defense confrontation,and analyzes the characteristics of off-line and off-line attack and defense in the line of sight.Firstly,a heuristic approximate dynamic programming based on action dependence is introduced to solve the problem of optimal control of single attack and defense against the existence of Nash-equilibrium solution in the line of sight.Secondly,for the optimal control problem of multi-body attack and defense in the line of sight,this chapter improves on the basis of the original natural operation algorithm,and introduces the “discount” factor to limit the influence range of future state on the current control quantity to reduce The solution amount increases the computational efficiency and is used to solve the optimal control instruction in the absence of the Nash equilibrium solution.Finally,for the multi-weapon platform cooperative attack and defense confrontation problem outside the line of sight,a cooperative guidance method based on the reference attack zone is proposed for the low-visuality target outside the line of sight,which is used to solve the optimal target for multiple out-of-sight targets.For the goal of high maneuverability outside the line of sight,a distributed cooperative guidance method with "Leader" mode is proposed based on the distributed communication topology.And the stability of the designed cooperative guidance law is proof in a mathematical form.In addition,the effectiveness and reliability of each method are verified by numerical simulation examples.