Research On Decision-Making Method Of Multi-Rotor UAV Manipulating Ground Dynamic Target

In recent years,the research on the autonomous execution of UAV in complex environment has attracted extensive attention in academic circles,especially the decision-making problem of UAV in dynamic and uncertain environment.This paper mainly focuses on the decision-making problem of rotor UAV in the process of ground dynamic targets,and mainly studies the subject from the following aspects.Firstly,optimize the design of UAV mission system structure.On the basis of comprehensive consideration of the environment of UAV maneuvering ground dynamic target mission,the system structure of the task is optimized.The designed system mainly includes four layers: perception,situation assessment,decision making and behavior control.Especially the situation assessment layer,which makes the structure of the UAV control mission system for the ground dynamic target more optimized,enables the UAV to complete the specified task in a more complex environment.Besides,The robustness of the system is improved.Secondly,the objective function is set and optimized according to the actual task,so that the UAV can select the ground multi-target in the dynamic environment.UAV can dynamically select the mission strategy in real time according to the designed strategy selection function,so that the UAV can select the dynamic target of the ground under the appropriate strategy.Based on NSGA-II,an algorithm for selecting multi-dynamic targets on the ground of UAV is designed.Thirdly,a decision-making method based on rough set theory for UAV to control dynamic target on the ground is designed.The key to maneuvering behavior of UAV to dynamic targets on the ground is accuracy and efficiency.The conditional attribute reduction method based on decision dependence is used to reduce the conditional attributes of the task,and the minimum decision table of the control behavior on ground targets of UAV is obtained.The minimum decision table is used to make decisions on the control behavior of UAV.Finally,this paper verifies that the multi-objective decision-making method of UAV based on NSGA-II algorithm is effective,and the decision-making method of UAV control behavior based on rough set theory is effective.By comparing with other decision-making methods,it is concluded that the decision-making method of UAV to ground dynamic target control task in this paper is more efficient.