Collaborative Task Assignment Methods For Multi-UCAV Ground Striking

With the development of modern cybernetics,artificial intelligence,information and communication technologies,the operational capability of the Unmanned Combat Aerial Vehicle(UCAV)has been greatly improved,which in the complex future battlefield can not only perform reconnaissance and early warning missions,but also perform highly coordinated missions,such as suppressing enemy air defense systems,striking enemy ground or sea targets,intercepting enemy tactical ballistic missiles and cruise missiles,even participating in air combat.In order to achieve the maximum operational benefit with the minimal operational cost,it is necessary to develop a reasonable mission plan for each UCAV(i.e.mission planning)with comprehensively considering the constraints of UCAV and weapon performance,the natural environment of the battlefield,etc.As a key part of mission planning,task assignment has an important impact on the rationality of planning results.To this end,this thesis focuses on the multi-UCAV ground-strike collaborative task assignment problem,constructs task assignment optimization models in different operational scenarios,and proposes corresponding solution methods,which can provide theoretical and methodological support for multi-UCAV mission planning.The main work and innovations of the thesis are as follows:(1)The collaborative task assignment problem of multi-UCAV ground-strike is analyzed.Firstly,the difference between task planning and task assignment is discriminated,and the research boundary of this paper is clarified.Secondly,based on the analysis of four key factors involved in the task assignment scheme,the collaborative task assignment process of multi-UCAV ground-strike is explained.Finally,three typical operational scenarios of multi-UCAV ground-strike are summarized,and the objective functions and constraints to be considered in each operational scenario are analyzed,which lays the foundation for subsequent research.(2)The task assignment problem of multi-UCAV collaborative striking intensive targets is studied.Aiming at the task assignment problem of multi-UCAV collaborative striking intensive targets,the relevant constraints are considered from the perspectives of combat coverage,combat feasibility and weapon resources,and an optimization model for maximizing the total damage expectation of enemy targets is established.The Hybrid Discrete Grey Wolf Optimization(HDGWO)algorithm is proposed to solve the model.Firstly,a decimal integer coding scheme is used to represent the grey wolf(the solution of the model),and a modular grey wolf position update method is proposed.Based on the above two mechanisms,it can be ensured that the constraint sets of the combat feasibility and weapon resources are always satisfied in the update process of population position.By introducing a penalty function into the objective function,it can be ensured that the grey wolf that violates the constraint set in terms of the combat coverage is continuously eliminated during the population position update process.On this basis,the Local Search(LS)algorithm is introduced to strengthen the development of the neighborhood space of the current optimal solution,so that the HDGWO algorithm has a good ability of global exploration and local exploitation.Compared with other typical algorithms,the proposed HDGWO algorithm is effective in solving the problem,which can not only obtain the global optimal striking scheme for small-scale formation in the shortest time,but also provide an approximate optimal solution for large-scale formation while achieving the best balance between solution quality and computation time.(3)The task assignment problem of multi-UCAV collaborative striking decentralized targets is studied.Aiming at the task assignment problem of multi-UCAV collaborative striking decentralized targets,the relevant constraints are considered from the perspectives of damage requirements,UCAV performance,UCAV operational requirements and variable range,and a multi-objective optimization model that minimizes total combat cost is established.In order to solve the model,the Three-phase Tabu Search(TTS)algorithm is proposed,including the weapon assignment phase,the initial solution generation phase,and the local search phase.In the weapon assignment phase,an optimization model consisting of partial objective functions and constraints is established to obtain the weapon assignment scheme that meets the target damage requirements and has the lowest cost.In the initial solution generation phase,considering that the weapon assignment scheme can be split,a splitting strategy based on the linear programming is proposed,based on which the initial solution generation methods based on distance clustering and random sequence are respectively provided.In the local search phase,six kinds of neighborhood search operators are put forward,and a quintuple-form tabu objective and a tabu list with variable length are designed,so that the tabu search can be executed.Simulation experiments show that the proposed TTS algorithm can effectively solve this kind of problem.(4)The task assignment problem of multi-UCAV collaborative striking decentralized targets under uncertain conditions is studied.In the modeling process,three uncertain factors are considered,including uncertain target information provided by multiple sensors,uncertain target weights,and partially known sensor weights.The ESMAA-2-ILP method,which combines the Extended Stochastic Multi-Criteria Acceptability Analysis-2(EMMAA-2)and the Integer Linear Programming(ILP),is proposed.Firstly,the uncertain target information provided by each sensor is represented as normal distribution interval numbers,and multiple decision matrices for each UCAV are calculated through the Weighted Arithmetic Averaging Operator(WAAO).Secondly,the final decision matrix is calculated by aggregating the above decision matrices through the iterative algorithm based matrix aggregation method.Then,based on the final decision matrix,the stochastic acceptability analysis is performed to calculate the Holistic Acceptability Index(HAI)of each UCAV against each target.Finally,based on HAI,the model containing uncertain parameters is converted into an integer linear programming model,so the task assignment scheme can be obtained through the integer linear programming method.Through simulation experiments,it is proved that the ESMAA-2-ILP method can effectively solve the task assignment problem of multi-UCAV collaborative striking decentralized targets with multiple uncertainties.