Research On Intelligent Monitoring System For Logistics UAV

Unmanned aerial vehicles are widely used in various fields,such as aerial photography,search and rescue,military reconnaissance,topographic survey,etc.,because of their flexible mobility and the advantages of not limited by geographical and environmental conditions.At the same time,the diversity and complexity of UAV application scenarios will inevitably lead to various abnormalities or faults of UAV components.It is essential for the UAV whether the drone components are in normal operation or not.As an important part of UAV system,UAV monitoring system is also facing the challenge of complexity and diversity of application scenarios.Therefore,it is of practical significance for the research and design of UAV monitoring system.Based on the application scenario of UAV logistics and transportation,this thesis designs an intelligent monitoring system for UAV,which has functions of front-end display,electronic map,route management,data storage and fault diagnosis.And it also introduces the function design and implementation of UAV motor fault diagnosis in detail.The monitoring system generally adopts a hierarchical structure design:communication layer,data layer,and application layer.Firstly,the wireless communication scheme between the UAV and the ground monitoring system is realized by using the principle and technology of serial communication.Secondly,the monitoring system uses the WPF interface framework under Windows to design the front-end interface of the monitoring system,and uses the C# language as the background logic code to build the software framework of the whole system.Then,the Google-Earth API interface is called to realize the functions of monitoring system electronic map navigation,aircraft track display and operation map based on the Java Script script.Then,the front-end data decoupling is complied,and the software design pattern is used to process the transmitted and received data,including displaying flight state data,transmitting control command data,operating route data,and data storage based on the MVVM data framework.Then,this thesis introduces the deep neural network algorithm and designs the fault diagnosis model.According to the problem of unbalanced fault samples,the SMOTE algorithm is employed to expand the data.According to the idea of integrated programming,the five monitoring quantities of the motor are adopted as the input of the fault diagnosis model.The monitoring and alarm function of the motor running state is realized on the monitoring system.When the monitoring system detects the fault,it will report the fault information to the UAV control system,so that the UAV control system can adjust the flight strategy and realize the intelligent monitoring function.Finally,the results of the actual flight test verify the effectiveness of the intelligent monitoring system of the logistics drone designed in this thesis.It is reliable and compatible,and improves the reliability of the UAV system.