Old buildings can now be made safe and collapse-proof using machine learning and artificial intelligence-aided mechanisms. Philadelphia-based Drexel University researchers have tried out a system to equip building inspectors with robotic assistants armed with AI and machine learning techniques to detect cracks in the innermost circuits of buildings, which can lead to major catastrophes if left undetected. The researchers have created autonomous systems to identify and inspect the problem areas.

The multi-scale system combines computer vision with a deep-learning algorithm. This pinpoints problem areas of cracking. It directs a series of laser scans of the regions to create a “digital twin” computer model. The system uses a high-resolution stereo-depth camera feed of the structure into a deep-learning program called a convolutional neural network. This assesses and monitors the damage. The process involves augmenting visual inspection technologies with a new machine learning approach.

The programs are in use in facial recognition, drug development and deepfake detection. They are now being adopted to spot patterns and discrepancies in volumes of construction data. The system represents a strategy to reduce inspection workload and help focus on preventing structural failures. The team is working on unmanned ground vehicles equipped with the newly-developed system to autonomously detect, analyze, and monitor cracks in structures.

They are creating a more intelligent, efficient system to maintain structural integrity across types of infrastructure. They are also looking at bring this newly reinforced technology to test in collaboration with the industry and regulatory bodies for practical applications.

Science is a world in itself. Here are some interesting facts that connect with you Nirad Mudur