摘要
针对以往使用的GPS、电涡流法、超声波法移动轨线自动识别系统受到噪声干扰,而导致识别精度低的问题,提出了基于深度八叉树的露天矿粘接机器人移动轨线自动识别系统设计。采用充电桩对电池进行充电,保证系统持续供电8 h以上。利用IMX179型照相机,从后方和侧面角度拍摄了露天矿粘接机器人的移动画面。利用摇把开关控制avt31 HD视频跟踪器,跟踪机器人的移动轨迹。采用深度八叉树将识别数据绘制到三维实体中,避免了繁琐的数据查找过程,方便了三维子体的定位。采用白平衡技术消除外界光线干扰,获得清晰的视频图像。构造深度八叉树存储结构,对其进行迭代处理,得到单通道亮度图像,即:轨道线。利用三阶贝兹曲线对其进行拟合,确定机器人的工作轨线。实验结果表明:系统轨线长度200 m,识别精准。
Aiming at the problem that the GPS,eddy current method and ultrasonic method mobile trajectory automatic identification system used in the past is interfered by noise,which leads to the problem of low identification accuracy,an automatic identification system design for the mobile trajectory of the bonding robot in open-pit mine based on the depth octree is proposed.Use the charging pile to charge the battery to ensure that the system can continue to supply power for more than 8 hours.Using the IMX179 camera,the moving pictures of the open-pit mine bonding robot were taken from the rear and side angles.Use the rocker switch to control the avt31 HD video tracker to track the movement of the robot.The recognition data is drawn into the three-dimensional entity by using the depth octree,which avoids the tedious data search process and facilitates the positioning of the three-dimensional sub-body.White balance technology is used to eliminate external light interference and obtain clear video images.Construct a deep octree storage structure,and iteratively process it to obtain a single-channel luminance image,that is,a track line.The third-order Bezier curve is used to fit it to determine the working trajectory of the robot.It can be seen from the experimental results that the trajectory length of the system is 200 m,which is consistent with the ideal situation and has accurate identification results.
作者
韩宝虎
赵亮
韩希君
HAN Baohu;ZHAO Liang;HAN Xijun(Storage Center of Hiller Energy Co.,Ltd.Hulun Buir 021000,Inner Mongolia China)
出处
《粘接》
CAS
2022年第3期172-176,共5页
Adhesion
基金
基于“大智物云”的焦化污染场地生物修复一体化智能装备研究(项目编号:2020YFC1807904)。
关键词
深度八叉树
露天矿
粘接机器人
移动轨线
自动识别
deep octree
open pit mine
bonding robot
moving trajectory
automatic identification
