7 Platform one US researchers are looking to use AI vision algorithms on autonomous vehicles to identify cracks in concrete (writes Nick Flaherty). The system, developed at Drexel University, combines computer vision from a stereo camera with a deeplearning algorithm to pinpoint problem areas of cracking before directing a series of laser scans of them to create a computer model, which can be used to assess and monitor the damage. “We aim to integrate this work with an uncrewed ground vehicle, enhancing the system’s ability to autonomously detect, analyse and monitor cracks,” said researcher Ali Ghadimzadeh Alamdari. “The goal is to create a more comprehensive, intelligent and efficient system for maintaining structural integrity across various types of infrastructure. Additionally, real-world testing, and collaboration with industry and regulatory bodies will be critical for practical application, and continuous improvement of the technology.” Training the algorithms on datasets of concrete structure images turns them into crack-spotters. “Taking a multi-scale, robotic approach will enable efficient pre-screening of problem areas via computer vision and precise robotic scanning of defects using non-destructive, laser-based scans,” said Dr Arvin Ebrahimkhanlou, an assistant professor on the project. “The neural network has been trained on a dataset of sample cracks. It can identify crack-like patterns in the images that the robotic system collects from the surface of a concrete structure. We call [these areas] ‘regions of interest’.” Once the ‘region of interest’ has been identified, the program directs a robotic arm to scan it with a laser line scanner, creating a 3D image. At the same time, a Lidar camera scans the structure surrounding the crack. Stitching both plots together creates a digital model of the whole area, showing the width and dimensions of the crack, and allowing tracking changes in between inspections. The team tested the system in the lab on a concrete slab with a variety of cracks and deterioration. The system was sensitive enough to pinpoint and accurately size up the smallest of fissures – less than a hundredth of a millimetre wide – outperforming top-of-the-line cameras, scanners and fibre-optic sensors. “This approach significantly reduces unnecessary data collection from areas that are in good structural condition while still providing comprehensive and reliable data necessary for condition assessment,” they said. Inspection equipment AI vision exposes cracks in concrete Uncrewed Systems Technology | April/May 2024 The system is sensitive enough to pinpoint and accurately size up the smallest of fissures – less than a hundredth of a millimetre wide The goal is to create a more comprehensive, intelligent and efficient system for maintaining structural integrity across various types of infrastructure
RkJQdWJsaXNoZXIy MjI2Mzk4