Research on Technology of Pipeline Detection Robot Based on Spiral Propulsion

• Published in: Intelligent Robotics and Applications Vol. 13016; pp. 67 - 78
• Main Authors: Wang, ZhiQian, Deng, ZhenBo, Lei, Pei
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer International Publishing AG 2021; Springer International Publishing
• Series: Lecture Notes in Computer Science
• Subjects: Pipeline robot; Point cloud processing; Roundness detection; Spiral propulsion
• ISBN: 3030890910; 9783030890919
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-030-89092-6_7

Aiming at the space curved pipeline with an inner diameter of less than Φ15 cm, in order to realize its internal detection work, this paper studies a compact, functionally integrated micro pipeline inspection robot and the corresponding pipeline inspection and scanning strategy. First, the robot adopts the driving principle based on spiral propulsion, and uses only one power source to realize the robot's forward, backward, and stop motions in the pipeline, and adapt to the bending and spatial posture of the pipeline. Ensure the robot's traveling ability and traveling efficiency in the pipeline. Second, the robot is equipped with a camera and a variety of measuring sensors to realize the detection of the inside of the pipeline. With the help of the robot's spiral propulsion action, the laser ranging sensor can scan the inside of the pipeline at 360°. And based on the robot's helix angle and coordinate conversion algorithm, the Cartesian coordinate system point cloud data set of the pipeline inner wall is obtained, and the point cloud data is processed based on the Poisson surface reconstruction algorithm to realize the visual surface reverse modeling of the pipeline inner wall, thereby Realize the foreign body inspection inside the pipeline. Third, this paper proposes a pipeline roundness detection algorithm based on point cloud data, which realizes the pipeline roundness detection. Fourth, this paper proposes a method for determining the location of foreign objects based on point cloud data. Finally, this paper designs a verification experiment to verify the robot's traveling mode and detection ability.