基于多重约束的铁路隧道净空检测研究

Research on Railway Tunnel Clearance Detection Based on Multiple Constraints

为解决铁路隧道洞内无法准确获取扫描仪定位参数和高精度三维点云数据成果难题,基于激光扫描仪、惯性测量单元、里程计等传感器,提出多源数据融合处理算法,获得三维点云相对数据成果;利用隧道自身的环缝、螺栓孔洞结构形状特征,生成点云灰度及深度影像,实现管片环缝提取,结合环片固定宽度完成里程校准;根据误差传播规律引入靶标控制点联合解算,修正定位定姿误差和坐标转换,获取工程坐标系统下的三维点云数据;基于最小二乘二维ICP匹配算法进行点云模板匹配,提取轨道中心线。经实验测试,对里程及靶标控制等多重约束后,获得的三维绝对点云平均误差为3.2 cm;提取隧道断面,输入隧道设计断面线并等分处理,与轨道基准坐标系进行匹配,逐点计算断面激光点到设计轮廓线的垂直距离,完成了隧道净空检测。实验数据分析表明,基于惯性测量单元/里程计组合导航误差纠正模型,在里程及靶标控制等多重约束后,能够获得高精度的铁路隧道三维点云数据成果,实现铁路隧道的净空检测。

In order to solve the problem that the scanner positioning parameters and high-precision three-dimensional point cloud data cannot be accurately obtained in the railway tunnel,a multi-source data fusion processing algorithm is proposed to obtain 3D point cloud relative data results based on sensors such as laser scanners,inertial measurement units,and odometers.Using the tunnels own circumferential seam and bolt hole structural shape features,the point cloud grayscale and depth images can be generated to extract the segment circumferential seam,and calibrate the mileage using the fixed width of the segment.According to the error propagation law,the target control point is introduced for joint calculation.Then the position and orientation errors are corrected and the coordinates are converted,after which the three-dimensional absolute point cloud under the engineering coordinate system is calculated.Based on the least squares two-dimensional ICP matching algorithm,the point cloud template matching is performed,and the track centerlines are extracted.Through experimental testing,the average error of the three-dimensional absolute point cloud obtained is 3.2 cm after multiple constraints of mileage and target control,.By extracting the tunnel section,inputting the tunnel design section line and dividing it equally,matching it with the track reference coordinate system,calculating the vertical distance from the section laser point to the design contour line point by point,the tunnel clearance detection is completed.The analysis of experimental data shows that,based on the IMU/odometer integrated navigation error correction model,it is possible to obtain high-precision 3D point cloud data results of railway tunnels and carry out the clearance detection of railway tunnels after applying multiple constraints of mileage and target control.