Pipeline Trajectory Reconstruction Based on Ensemble Empirical Mode Decomposition With Partial Adaptive Noise

In recent years, pipeline inspection and maintenance play an important role in the field of infrastructure management. One of the key challenges in pipeline inspection involves the precise determination of the pipeline trajectory, which is critical for identifying potential issues such as leak and corrosion. Inertial sensors, such as accelerometers and gyroscopes, are commonly used to track the movement of the wheel-type pipeline inspection robot (WPIR) and reconstruct the pipeline trajectory. However, the complete inertial measurement unit (IMU) possesses the disadvantages of high cost and complexity, which limit the practical application. Therefore, in this article, we propose a reduced inertial sensor system (RISS) that only requires a single-axis angular velocity meter and a dual-axis accelerometer. An integrated empirical mode decomposition (EMD) method with partial adaptive noise (AN) is investigated to improve the accuracy of sensor data acquisition. The proposed method has been validated through simulations and real tests, and it shows promising results for accurately reconstructing underground pipeline trajectories with lower computational effort and higher confidence in higher order components.