Research On Task Allocation Technology Of Multi-UAV Cooperative Combat

In recent years,unmanned aerial vehicle(UAV)has shown enormous potential in the military field due to their advantages such as small size,low cost,high flexibility and high maneuverability.Multi-UAV cooperative operation can enhance the flexibility,reliability and real-time performance of the system,which has become a research hotspot in recent years.The task assignment scheme of multi-UAV has an important influence on the combat effectiveness of the system.Due to the constraints of UAV’s own capability,target characteristics and battlefield environment,the task assignment problem of UAV can be regarded as an NP-difficult combinatorial optimization problem.In this thesis,under the background of multi-UAV cooperating with each other to perform reconnaissance,strike and evaluation tasks on targets,the path planning problem of UAVs,task pre-assignment problem in static environment and real-time task reallocation problem in dynamic environment are studied.The main research contents are as follows:Firstly,the problem of UAV path planning is studied.A path planning model for UAVs is established.The model takes the shortest path as the goal,and mainly considers the environmental constraints such as terrain,threat area and the performance constraints of UAVs.BASGA algorithm was proposed to solve the model,and the BAS(Beetle Antennae Search)algorithm is pulled into the genetic algorithm to solve the defect of genetic algorithm easy to fall into the local best.Simulation results prove that the BASGA algorithm can obtain a shorter feasible path.Then the multi-UAV pre-allocation problem is studied.The model of multi-UAV task pre-allocation problem is established.The model aims at maximizing profit,minimizing range and time cost,and mainly considers constraints such as executive ability,resources,maximum range and mission timing.An improved adaptive genetic algorithm based on multiple types of genes was proposed.The algorithm encoded for the heterogeneous characteristics of UAV,and the initialization,selection,crossover and mutation operations were redesigned.The adaptive crossover rate,mutation rate and population updating mechanism are proposed.Simulation results prove that the improved algorithm can get a better solution.Finally,the multi-UAV redistribution problem is studied.The emergent situations that need dynamic redistribution are analyzed,and the task redistribution model of multi-UAV is established.This thesis proposes an improved contract net algorithm,which uses three types of contracts:buying and selling,exchange and replacement,introduces management UAV,sets bidding threshold,introduces concurrency mechanism,and considers various solutions to conflicts.Simulation results prove that the algorithm proposed in this thesis has the advantages of less traffic,higher real-time performance and higher reliability.