A Simultaneous Pipe-Attribute and PIG-Pose Estimation (SPPE) Using 3-D Point Cloud in Compressible Gas Pipelines

Area of Science:

• Engineering
• Robotics
• Geophysics

Background:

• Accurate in-line inspection (ILI) of underground gas pipelines requires precise knowledge of pipe attributes and the pose of pipeline inspection gauges (PIGs).
• Existing methods may lack the accuracy or efficiency needed for complex pipeline networks.

Purpose of the Study:

• To develop a simultaneous pipe-attribute and PIG-pose estimation (SPPE) approach for enhanced ILI.
• To accurately estimate pipe dimensions, PIG orientation (roll, pitch, yaw), and position relative to the pipe center.

Main Methods:

• Utilizing 3D point cloud data from LiDAR or ToF cameras as input.
• Modeling the problem as an optimal transformation matrix estimation from PIG sensor frame to global pipe frame.
• Decomposing the transformation into two sub-transformations, including non-linear optimization with the Levenberg-Marquardt algorithm (LMA) and gravity vector utilization.

Main Results:

• The SPPE approach accurately estimates pipe attributes and PIG pose.
• Simulations on a ROS-based PIG simulator demonstrate excellent accuracy.
• The method is computationally efficient, suitable for real-time and post-processing NDT applications.

Conclusions:

• The proposed SPPE method offers a significant advancement in pipeline inspection technology.
• Its accuracy and efficiency make it valuable for ensuring the integrity of underground gas pipelines.
• SPPE is applicable to both real-time and post-processing non-destructive testing scenarios.