高级检索+

猪舍消杀巡检机器人系统设计与试验

Research and Design of Pig House Disinfecting Inspection Robot System

  • 摘要: 针对规模化猪舍人工劳动强度大、重复作业多、疫病传播与防控形势严峻等问题,设计了猪舍消杀巡检机器人系统。该系统融合基于2D激光雷达的即时定位与建图(Simultaneous localization and mapping, SLAM)和超宽带(Ultra-wide band, UWB)技术,实现舍内地图构建和系统实时定位;在确定热红外模组安装高度为125 cm和安装倾角水平向下夹角5°的基础上,运用Jetson Xavier NX边缘计算单元进行视觉处理与识别算法的部署,完成在线猪只体温巡检;边缘计算单元依据终端指令对消杀模块中超声波雾化单元、紫外线辐射单元等进行决策控制,实现多模式舍内环境消杀;通过传感器技术对舍内环境参数进行实时监测;并搭建人机交互界面,实现监测信息的显示、报警、存储等。测试结果表明,该系统可完成地图构建、自动导航、猪只体温检测,记录异常猪只热红外图像及圈舍所在位置;依据设定的消杀模式,在目标点开启相应消杀功能的准确率为100%;机器人在巡检状态和静止状态下,舍内CO2浓度、温度、相对湿度的相对误差分别为0.04%、3.00%、2.10%。本研究可为疫情形势下猪舍巡检消杀少人化/无人化作业提供技术装备参考。

     

    Abstract: In order to solve the problems of large-scale pig house, such as high labor intensity, repeated operations and severe epidemic situation, a pig house disinfection and sterilization inspection and environmental monitoring systemwas designed.The system integrated the 2 D LiDAR based simultaneous localization and mapping(SLAM) and ultra wide band(UWB) technologies to realize the building of in-house map and real-time positioning of the system. On the basis of determining the installation height of the thermal infrared module to be 125 cm and the horizontal downward angle of the installation inclination to be 5°, the Jetson Xavier NX edge computing unit was used to deploy vision processing and recognition algorithms to conduct online inspections of pig body temperature.The edge computing unit was used to intelligently control the atomization module and the ultraviolet module in the disinfection module according to the terminal instructions and the information of the micro-environment in the house, so as to realize the multi-mode disinfection of the indoor environment.The sensor technology was used to monitor the environmental parameters in the house in real time.Human-computer interaction interface was built to realize the display, alarm and storage of monitoring information.The test results showed that the system can complete map construction, automatic navigation, pig body temperature detection, and record abnormal pig thermal infrared pictures and the location of the enclosure.According to the set disinfection and sterilization mode, the accuracy of completing the corresponding disinfection and sterilization function at the target point was 100%.The relative errors of CO2 concentration, temperature and humidity in the room were 0.04%, 3.00% and 2.10% respectively in the inspection state and static state of the robot. The research result can provide technical equipment reference for the less-humanized/unmanned operation of inspection and disinfection of pig houses during the epidemic situation.

     

/

返回文章
返回