Abstract
The inspection of iron and steel pipes in industrial settings is a critical aspect of ensuring structural integrity and preventing potentially catastrophic failures. Traditional inspection methods have limitations in terms of accessibility and speed. This paper presents a novel solution in the form of a caterpillar robot designed for inline inspection of iron and steel pipes. The caterpillar robot's design, integration of advanced sensor technology, navigation, and control systems are described in detail. The paper also discusses the algorithm used for corrosion and crack detection. Results from comprehensive testing demonstrate the robot's ability to detect defects with high accuracy and reliability, thereby improving maintenance and repair decisions. The caterpillar robot offers a promising alternative to traditional inspection methods, with potential applications in various industrial contexts. This paper sheds light on the development and potential adoption of caterpillar robots for enhanced pipe inspection in iron and steel industries. The heart of the system lies in the integration of image processing techniques. The robot employs machine vision, enabling the detection and precise identification of defects within the pipe's inner surface. A prototype model is developed, featuring caterpillar-like legs driven by electromagnetic mechanisms, allowing it to move efficiently within the pipe. The robot's image processing capabilities are realized using a Raspberry Pi camera The image processing aspect of the project involves a range of transformations,including image blurring, smoothing, binary conversion, canny edge detection, logarithmic transformations, and contouring. These techniques are used to process captured images, providing valuable insights into the condition of the pipe. The results obtained from this research are instrumental in assessing the current state of the pipe, aiding in the early detection of potential issues, and thereby preventing costly and disruptive pipe failures Keywords: In-Pipe Inspection Robots, Raspberry Pi camera, Image processing
Authors
