Research On Planning And Execution System Of Farmland Data Acquisition Task Based On UAV

As an indispensable intelligent device in smart agriculture,UAV can carry out efficient farmland data collection operations and crop growth management and monitoring,which is an important means of farmland task execution.At present,there are some problems in UAV farmland data acquisition,such as unreasonable path planning,low working efficiency and data acquisition efficiency,and untimely data processing.Therefore,in order to improve the execution efficiency of farmland tasks and solve the problem of untimely data processing,this paper constructs the farmland UAV path planning scheme by designing the navigation point path optimization algorithm,the area coverage algorithm and the area sampling algorithm.On this basis,the parallel transmission technology is used to design the UAV data transmission and processing scheme.Finally,the UAV based farmland data acquisition task planning and execution system is developed by using the Android development technology,Flask framework technology and front-end vue technology.The main research results are as follows:1.Build UAV path planning algorithm.In order to improve the efficiency of UAV farmland image data acquisition,a UAV path planning algorithm based on improved genetic algorithm is proposed,which is suitable for UAV low-altitude image data shooting in farmland environment.By comparing the improved genetic algorithm with the original genetic algorithm,the computing speed is improved by 76.73%,and the accuracy of the results is also improved.In this paper,an adaptive large neighborhood search algorithm with multiple heuristic factors is selected for comparison.The results show that the improved genetic algorithm can get the results faster,and with the increase of the number of navigation points,the adaptive large neighborhood search algorithm has higher solving accuracy.Then,the algorithm is applied to the system.By comparing the flight time of the optimized navigation point mission with the flight time of the pre-optimized navigation point mission,the time efficiency increases by 15.68% on average.The results show that the improved genetic algorithm is optimal both in terms of speed and accuracy when the number of navigation points is small.When the number of navigation points is large,the adaptive large neighborhood search algorithm is a better choice.2.Uav data parallel transmission and processing scheme.In order to improve the transmission rate between the UAV image data and the server,multithreading is integrated into each channel on the basis of multi-channel to achieve parallel channels and parallel within a single channel.In this study,60-300 UAV images were selected to carry out data transmission tests on single channel,double channel,three-channel and four-channel respectively.It was found that in the whole experiment process,the network bandwidth of the transmission terminal did not increase due to the increase of transmission channels.In all tests,the network transmission speed was about 11-12MB/s.Although limited by the network bandwidth of the transmission end,the transmission efficiency of UAV image data is also improved when the channel increases slowly.When the amount of UAV image data remains unchanged,the transmission channel is increased to three channels,and the data transmission efficiency is increased by about 22.5%;when the transmission channel is increased to four channels,the data transmission efficiency is increased by about 24%.When the transmission channel is increased to five channels,the data transmission efficiency is still improved by about 24%.With the increase of the number of channels,the data transmission efficiency is greatly improved when it is increased from single channel to four channels,and is basically the same as that of four channels when it is increased to five channels.The results show that the bandwidth utilization of the transmission terminal can be effectively improved by increasing the number of channels,but the utilization decreases with the increase of the number of channels.Therefore,4 parallel channels are the most cost-effective.3.The planning and execution system of farmland data acquisition task based on UAV is constructed.Based on DJI UAV equipment,it uses DJI MSDK to acquire the control flight of UAV,etc.,and realizes the acquisition,uploading and processing of the data collected by UAV.The UAV mobile ground station subsystem is formed.In order to facilitate the production of route tasks by users,the UAV route task package production subsystem is developed to make the region of the task packages produced by users more accurate.In order to facilitate the task package making users to obtain the execution state of UAV route flight task,a UAV route task visualization subsystem was developed.The three subsystems are integrated to form a UAV-based farmland data acquisition task planning and execution system.