Measurement of long-distance buried pipeline centerline based on multi-sensor data fusion

This study presents a multi-sensor data fusion method for solving the problem in surveying long-distance oil-gas pipeline centerline. During in-line inspection (ILI) of long-distance oil-gas pipeline centerline, an inertial measurement unit (IMU) and an odometer are loaded in the ILI. Additionally, a low-frequency transmitter is loaded to activate the global positioning system (GPS) marker placed on the ground which detects the location of ILI. An error model is constructed using the inertia navigation system (INS) and nonlinear dynamic system. Dead reckoning is performed with IMU data and odometer readings. All errors of the INS and odometer are estimated and compensated using extended Kalman filter. Finally, accurate position of pipeline centerline is corrected with data of the GPS marker. Trajectory curves of the same pipeline centerline are compared between ILI surveys at two different times. Results show that the integrated errors of INS-based centerline data are 0.35 m in the horizon and 0.74 m in the height. Accuracy of the ILI is verified by excavating a buried pipeline at one feature point, demonstrating an ILI error