摘要
为了解决液压支架护帮板在收回过程中的空间姿态测量问题,提出了一种基于深度视觉原理的多传感器融合护帮板空间姿态测量解决方案。该方案通过FAST算法提取特征点,采用领域搜索算法优化特征点,基于护帮板姿态解算模型采用深度相机和倾角传感器相融合的方式进行姿态解算。结果表明:特征点在经过模型优化后,准确度提升25%,液压支架护帮板偏航角的平均解算误差为0.68°,横滚角的平均解算误差为1.31°,俯仰角的平均解算误差为0.82°,护帮板空间姿态角度解算最大误差为1.91°,护帮板空间姿态角度解算最小误差为0.57°,满足井下护帮板姿态检测要求。该方法不易受到环境的干扰,便于获取护帮板在三维空间中的姿态,不但可用于液压支架护帮板空间姿态测量,而且在其他领域视觉测量技术中也具有较好的普适性。
In order to solve the problem of space attitude measurement of hydraulic support side guard in the process of retraction,a multi-sensor fusion solution of space attitude measurement of side guard based on depth vision principle was proposed.The scheme extracts feature points by FAST algorithm,optimizes feature points by domain search algorithm,and calculates the attitude based on the attitude calculation model of the side guard by integrating depth camera and tilt sensor.The results show that after the optimization of the feature points,the accuracy increases by 25%,the average measurement error of the side guard yaw angle of the hydraulic support is 0.68°,the average measurement error of the roll angle is 1.31°,the average measurement error of the pitch angle is 0.82°,the maximum error of the side guard spatial attitude angle solution is 1.91°,and the minimum error of the side guard spatial attitude angle solution is 0.57°,which meets the requirements of the side guard's attitude monitoring of the hydraulic support.This method is not easy to be disturbed by the environment,and it is convenient to obtain the attitude of the side guard in three-dimensional space.It can not only be used for the measurement of the spatial attitude of the hydraulic support side guard,but also has good universality in visual measurement technology of other fields.
作者
张丹
陈仕林
吴卫东
宋胜伟
李士魁
Zhang Dan;Chen Shilin;Wu Weidong;Song Shengwei;Li Shikui(School of Mechanical Engineering,Heilongjiang University of Science and Technology,Harbin 150022,China;Shuangyashan Shuangmei Electromechanical Equipment Co.,Ltd.,Shuangyashan 155110,China)
出处
《煤矿机械》
2023年第9期191-194,共4页
Coal Mine Machinery
基金
黑龙江省首批揭榜挂帅项目(2021ZXJ02A01)。
关键词
深度视觉技术
FAST特征点算法
领域搜索算法
液压支架护帮板
machine vision technology
FAST feature point algorithm
domain search algorithm
hydraulic support side guard