The invention discloses an underground monorail hoist unmanned driving system based on machine vision, which comprises a vehicle-mounted camera for acquiring road condition video data when a monorail hoist runs, a millimeter-wave radar for measuring distance of obstacles encountered when the monorail hoist runs, a vehicle-mounted processor for generating a region of interest ROI after radar data conversion, and a vehicle-mounted camera for acquiring the road condition video data when the monorail hoist runs. Inputting the road condition video data into an improved TinyYOLOv3 neural network to obtain an ROI (Region of Interest) processed by the neural network, and associating and matching the ROI converted from the radar data with the ROI processed by the neural network to obtain the type of an obstacle, the position of the obstacle and the relative speed of the obstacle and the monorail crane. According to the method, image recognition and information fusion are achieved on the monitoring front end, so that the calculation pressure of the monitoring platform server is relieved.
本发明公开了一种基于机器视觉的井下单轨吊车无人驾驶系统,包括:车载摄像头采集单轨吊车行车时的路况视频数据,毫米波雷达对单轨吊车行车时遇到的障碍物测距,车载处理器生成雷达数据转换后的感兴趣区域ROI,将路况视频数据输入改进的TinyYOLOv3神经网络,得到神经网络处理后的感兴趣区域ROI,将雷达数据转换后的感兴趣区域ROI和神经网络处理后的感兴趣区域ROI关联匹配,得到得到障碍物类型、障碍物的位置和障碍物与单轨吊的相对速度,输入远程监控子系统生成控制指令,远程控制单轨吊车进行行车,该方法监控前端上实现图像识别和信息融合,以减轻监控平台服务器的计算压力。
Underground monorail hoist unmanned driving system and method based on machine vision
一种基于机器视觉的井下单轨吊车无人驾驶系统和方法
29.07.2022
Patent
Elektronische Ressource
Chinesisch
Underground monorail hoist convenient for visual positioning
Europäisches Patentamt | 2024
|Underground pneumatic monorail hoist and intelligent control system thereof
Europäisches Patentamt | 2024
|