Modeling Research On UAV Operations Simulation Training System

Plant protection is an important scenario for UAV applications and is currently an important direction for UAV vocational education and industry training in China.Adequate hands-on training is the only way to ensure the standard of plant protection operators.However,due to the tight civil airspace in China and the limitations of the teaching space,as well as safety considerations,operational training in this area can only rely on simulation systems,so drone simulation training systems have become an important research topic.In this paper,the optimal decision kernel for the simulation training system of plant protection UAV operation is studied algorithmically and the simulation system is developed with UAV as the research object and agriculture and forestry plant protection as the application scenario.The operational scenarios of the training system include irregular operational areas and irregular and discontinuous area operations.The user makes the trajectory planning decision in these two scenarios through the simulation system;and the simulation system needs to have an embedded optimal decision kernel to compare the trainee’s decision solution with the optimal result in this scenario in order to improve the rationality of its decision,so this paper focuses on the following three aspects.1.Single plant protection UAV operation for irregular operation areas.Firstly,the operating environment is modeled using the introduced black zone raster method based on binary expansion,then the genetic algorithm is used to solve the problem,and finally the applicability of the proposed algorithm to path planning in irregular operation area is verified by experiments.2.This paper converts the multi-plant protection UAV cooperative operation in a discontinuous operation area into an MTSP problem,which is solved using a genetic algorithm.Since the genetic algorithm has the limitation of specific codes for specific task sets when solving the MTSP model,this problem is solved by combining the genetic algorithm with the K-means algorithm for the optimization,and the correctness of the collaborative optimization algorithm combining the genetic algorithm with the K-means algorithm is verified through experiments.3.Based on the above research work,this paper adopts Java EE technology,Agent technology and SSM framework,in Intelli J IDEA,Anylogic platform,to develop the system,including vision simulation,path planning,task training,assessment and other modules,to achieve the job scene display,optimal path decision,user interaction,assessment results display,data storage and other functions The system has been evaluated and tested for use in training.After evaluation and testing the system can be used for training,which is supportive for vocational education and industry training of UAV plant protection.