测试技术学报2024，Vol.38Issue(4)：427-434,8.DOI:10.3969/j.issn.1671-7449.2024055
基于机器视觉的装车臂自动对接系统研究
Research on Automatic Docking System of LNG Land Loading Arm Based on Machine Vision
孙泽昆 1陆永华 1朱赟 2冯是公 3秦栋成3
作者信息
- 1. 南京航空航天大学 机电学院,江苏 南京 210016
- 2. 中国航发控制系统研究所,江苏 无锡 214063
- 3. 江苏长隆石化装备有限公司,江苏 苏州 215600
- 折叠
摘要
Abstract
Aiming at the problems of manual pushing,difficult operation and low work efficiency of load-ing arm docking,a machine vision-based automatic docking system of LNG land loading arm is designed and studied,including target detection,positioning and control methods for automatic docking in the delim-ited parking area to realize intelligent docking in unmanned state.The first is to control the joint angle through the servo motor position ring and record the rotation position;Secondly,based on the geometric projection method and the structural characteristics of the loading arm,the kinematics algorithm is simpli-fied to improve the calculation efficiency,and the kinematic model is established by the D-H method,the motion range is calculated and the geometric projection method is verified.Subsequently,object detection is realized based on deep learning and image processing algorithms.Finally,according to the camera inter-nal parameters,multi-point depth information and target detection results obtained by camera's calibra-tion,three-dimensional vision is realized through the principle of monocular positioning,and finally auto-matic docking is realized.The test of the flange in the parking area of the vehicle shows that the success rate of the system docking is 99%,and the time is shortened by 40%compared with the manual docking.The automatic docking system of loading arm can be effectively applied to the loading and unloading truck work of the LNG receiving station,and has certain reference value for realizing automatic docking.关键词
LNG陆用装车臂/机器人运动学/机器视觉/目标检测/自动对接Key words
LNG land loading arm/Robot kinematics/machine vision/object detection/automatic docking分类
信息技术与安全科学引用本文复制引用
孙泽昆,陆永华,朱赟,冯是公,秦栋成..基于机器视觉的装车臂自动对接系统研究[J].测试技术学报,2024,38(4):427-434,8.
