Abstract
In today's modern world, pipelines serve as a vital component of the economic infrastructure of countries, serving as a lifeblood vascular system. To prevent disasters, timely inspection and maintenance operations are critical. In-pipe inspection robots are emerging as a promising solution to prevent any negative consequences in the pipeline industry. In this research, we propose a new design concept of motor placement and extension system that employs compression springs as the passive extension mechanism. The primary objective of the new design is to provide video feedback, enabling the robot to detect and recognize various types of obstructions and leak sources, including cracks, corrosion, weld defects, and deformations of the pipe structure. To further enhance the capabilities of the robot, a deep learning system will be applied with a carefully created dataset. This will enable the robot to learn and classify anomalies with increased accuracy and efficiency. This research paper is organized in the following order; introduction, robot design, control board, defect detection system, and results part with evaluation.