基于嵌入式互联网的远程智能喷雾控制系统设计

Design and experiment of remote intelligent spray control system based on embedded internet

为提高设施农业植保作业智能管控能力,该文提出一种基于STM32F101和STM32F103嵌入式技术,结合4G互联网、局域WIFI通信技术及超声波靶标检测算法,能够便捷地对设施作业装备远程视频与控制的方案,达到人机分离与精准施药的目的。该系统在Eclipse和Keil-uvision4开发环境下采用Socket和多线程技术实现双向通信,以TCP通讯协议为媒介,Android端和客户端通过互联网或无线网卡转接移动路由实现远程智能喷雾控制。试验结果表明:1)Android端能够在LAN或Internet中实现智能喷雾装备的近远程控制,软件界面回传状态无卡顿、延时发生,能够准确发射控制指令,实现了对靶标间歇性施药管控;2)系统建立的双向心跳包能够在通信故障情况下迫使喷雾装备处于休眠状态,经测试,心跳包设定时间与喷雾装备休眠响应时间平均相对误差不超过5.50%;3)采用视频帧对冠层中线定位,借助超声检测算法确定风送距离参数且建立冠层体积模型。试验发现,冠层密度对体积测量结果有显著影响,但总体测量准确度可达94.67%。该研究对其他农机装备的智能化管控研究有参考意义。

In view of the current weak status of intelligent management and control systems for facility of agriculture and dwarfed orchards,high labor intensity and complicated spraying environment,this article proposed an embedded technology,which took STM32F101 and STM32F103 as the core and basis,combined with 4G Internet and ultrasonic target detection algorithm,and was able to remotely control agricultural machinery to achieve the purpose of human-machine separation and precise application of pesticides.The system used Socket and multi-thread technology to implement two-way communication under the Eclipse and Keil-uvision4 development environment.With TCP communication protocol as the medium,the Android and client transferred remote routing through the Internet or wireless network card to achieve remote intelligent spray control.The test results show that:1)Android can implement near-remote control of smart spray equipment in the LAN or Internet.The software interface returns that no stagnant state or delay occurs and it can accurately transmit control commands and achieve intermittent application of the target.2)The bidirectional heartbeat package established by the system can force the spray equipment to be dormant in the event of communication failure.After testing,the error rate of the heartbeat packet set-up time and the spray equipment dormancy response time does not exceed 5.50%.3)Use the video frame to locate the canopy middle line,and use the ultrasonic detection algorithm to determine the wind-distance parameters and establish the canopy volume model.Through experiments,it is found that the degree of canopy denseness has a significant impact on the measured results of the canopy volume measurement system.The text measurement system should be used in dense canopy conditions,and the measurement accuracy is as high as 94.67%.4)The remote smart spray control system in this article can realize local network and long-distance internet communication.Use“Ping”command to check the network response to IP data packets.There is no packet loss problem and the average response time is 5 and 79 ms,respectively.5)The degree of canopy density has a significant effect on the measurement of the canopy volume measurement system.There is a significant difference between the manual and automatic measurements of the canopy volume in sparse conditions.On the contrary,there is no significant difference between the manual and automatic measurements of the canopy volume in the dense conditions,which is suitable for measuring the canopy volume in this article.In the system,the maximum relative error is only 6.4%,and the canopy coverage and distribution uniformity will be significantly improved.6)A remote control and optimized application system available for complete Internet-based spray equipment is designed.The system can realize human-machine separation and precise application of pesticides.Through tests,it can be seen that the detection and processing of the fuselage faults has high reliability.At the same time,it has potential application value for the transformation of traditional production modes and the promotion of smart agricultural machinery.The study can also provide reference for the automation and intellectualization of other agricultural machinery equipment.