Abstract—

To produce a 3D map of the Tehran’s first gas transfer pipeline (Tehran—Kuhnamak), a methodology has been developed in this research, in which a strapdown inertial navigation system (SINS) based on micro-electro-mechanical system (MEMS) and inertial measurement unit (IMU) is applied on pipeline inspection gauges (PIGs) to sense data every 4 millimeters of 111 kilometers of the whole pipeline. The navigation solution is based on an extended Kalman filter (EKF) using Allan variance (AVAR) to analyze and tune the EKF initial inputs. A new constrained PIG filter (CPF) is proposed in this paper in integration with EKF, in which two Euler angles (pitch and yaw) of the PIG are updated due to non-holonomic state constraints between pipe junctions. Besides, 98 magnetic control points have been used to increase robustness about every kilometer, which is coordinated by GPS. Furthermore, odometer measurements have been employed as measurements in the EKF. The results show that using such a hybrid approach has improved the PIG positioning accuracy by about 81% compared with that of the Basic EKF. In addition, positioning accuracy in comparison with the latest methods like EKF/pipeline junctions (PLJ) has increased by 32%. Furthermore, the proposed method is 55% better than EKF/PLJ in the algorithm runtime.

中文翻译：

---

一种新型约束滤波器与扩展卡尔曼滤波器集成在使用 MEMS IMU 的地下管道导航中

摘要——

为了制作德黑兰第一条输气管道（德黑兰-库纳马克）的 3D 地图，本研究开发了一种方法，其中基于微机电系统 (MEMS) 和惯性的捷联惯性导航系统 (SINS)测量单元 (IMU) 应用在管道检测仪表 (PIG) 上，以在整个 111 公里的管道中每 4 毫米检测一次数据。导航解决方案基于使用艾伦方差 (AVAR) 的扩展卡尔曼滤波器 (EKF) 来分析和调整 EKF 初始输入。本文提出了一种与 EKF 集成的新约束 PIG 滤波器（CPF），其中 PIG 的两个欧拉角（俯仰角和偏航角）由于管接头之间的非完整状态约束而被更新。此外，还使用了 98 个磁控点来增加每公里的鲁棒性，由 GPS 协调。此外，里程计测量已被用作 EKF 中的测量。结果表明，与Basic EKF相比，使用这种混合方法将PIG定位精度提高了约81%。此外，与 EKF/管道接头 (PLJ) 等最新方法相比，定位精度提高了 32%。此外，所提出的方法在算法运行时比 EKF/PLJ 好 55%。