Research On UAV Reconnaissance Mission Planning Algorithm Based On HTN

With the increasing progress of science and technology,unmanned aerial vehicle technology is gradually becoming mature.Lightweight and replaceable distributed unmanned aerial vehicle cluster has a good application prospect in both military and civilian fields.How to efficiently plan unmanned aerial vehicle mission has increasingly become a common key issue in related research fields.Layered task network HTN(Hierarchical Task Network,HTN)has a wide range of applications in the field of mission planning.Most of the current research focuses on civil fields such as logistics and transportation,and is rarely used in reconnaissance and surveillance task scenarios.However,the intuitive and concise decomposition of task planning by HTN is urgently needed in UAV reconnaissance task planning.Therefore,this thesis will study the application of HTN technology in UAV reconnaissance mission planning,and the task assignment and route planning in UAV mission planning process are studied and simulated.The main research contents of this thesis are as follows:(1)Uav reconnaissance mission planning is realized by HTN planning technologyIn this thesis,the task planning problem of UAV reconnaissance scene based on HTN is deeply studied.By using JSHOP2 planner of HTN hierarchical task network technology,the simple UAV reconnaissance process is defined,decomposed and planned.On this basis,single task is extended to multi-task and single region is extended to multi-region.The simulation results show that HTN can accomplish a lot of planning through the design of decomposition unit and the definition of domain problems,and it can also have good application effect when the number of task objectives and execution objectives is large.(2)Uav reconnaissance and positioning task assignment based on improved ant colony algorithmThe target position should be obtained before the reconnaissance mission.In this thesis,the Chan-Taylor joint positioning algorithm is used to study the influence of the number of UAVs and baseline length on the positioning algorithm.On the basis of high positioning level,after comparing the advantages and disadvantages of the UAV reconnaissance task assignment based on HTN and ant colony algorithm,the ant colony algorithm is used to realize the UAV task assignment process,and the pheromone updating strategy of the ant colony algorithm is improved.The experimental results show that,The probability of the improved ant colony algorithm getting task assignment results within 10 iterations is increased by 12%,the probability of getting task assignment results within 20 iterations is increased by 6%,and 100% can get task assignment results within 20 iterations.At the same time,the time required by the ant colony algorithm and the improved ant colony algorithm to obtain task assignment results in the reconnaissance scenario is compared with the Hungarian algorithm commonly used in task assignment.The experimental results show that the improved ant colony algorithm can complete the UAV reconnaissance task assignment faster in the reconnaissance scenario.(3)Uav reconnaissance route planning based on improved particle swarm optimization algorithmThe resource consumption of UAV is subject to coupling constraints of various factors and has different solution sets under different simulation conditions.After comparing the UAV reconnaissance route planning problems based on HTN and particle swarm optimization algorithm,this thesis conducts mathematical modeling for UAV reconnaissance route planning and uses improved particle swarm optimization algorithm to realize the route planning process.In the same reconnaissance scenario,The fitness values of the improved particle swarm optimization algorithm are compared with those of particle swarm optimization algorithm and immune algorithm.The experimental results show that the fitness value of the improved particle swarm optimization algorithm is higher and higher with the increase of the number of iterations,and the improved particle swarm optimization algorithm can solve the problem that the particle swarm optimization algorithm is easy to fall into the local optimal solution and the algorithm convergence is too slow in the late stage in the UAV reconnaissance route planning.In this thesis,a simulation experiment is carried out to verify the problem of UAV reconnaissance mission planning under the reconnaissance scenario.The experimental results show that HTN is suitable for UAV reconnaissance mission planning.At the same time,the improved ant colony algorithm to achieve UAV reconnaissance task assignment can obtain the task assignment result in fewer iterations,and the improved particle swarm optimization algorithm to achieve UAV reconnaissance route planning can effectively avoid obstacles and obtain the UAV route correctly.