针对小口径管道内检测场景，提出了基于MEMS惯导系统的小口径油气管道中心线测绘技术，自研了基于捷联惯导平台的速度、位姿与位置解算算法。根据捷联惯导、轮式里程计的误差方程，使用卡尔曼滤波器对IMU输出信息进行补偿矫正，最后融合地表磁标记信息获得高精度的管道中心线位置信息。利用某内检测项目的实际检测数据进行算法效果验证，并对比了纯惯导方案与多传感器融合方案的效果。结果表明，提出的导航算法可以将检测精度控制在每公里1 m以内，改善了纯惯导测量的检测效果，满足工程应用对检测精度的要求。

Aiming at the internal detection scene in small-diameter pipelines, this paper proposed a small-diameter oil and gas pipeline centerline mapping technology based on MEMS inertial navigation system. The algorithm of velocity, pose and position solution based on strapdown inertial navigation system was developed as well. According to the error equations of strapdown inertial navigation and wheel odometer, Kalman filter was used to compensate and correct the output information of IMU. Finally the high-precision pipeline centerline position information was obtained by fusing ground magnetic marking information. In this paper, the effectiveness of the algorithm was verified by the actual test data of an internal detection project. The simple inertial navigation scheme and the multi-sensor fusion scheme were also compared. The experimental results indicated that the proposed algorithm can control the detection precision within 1 meter per kilometer, which greatly improved the detection ability of simple inertial navigation measurement to meet the detection accuracy of practical engineering applications.