An AI program is being developed that, when combined with footage from unmanned aerial vehicles (UAVs), could be used to conduct automated investigations of building enclosures, possibly identifying problems earlier. Thornton Tomasetti Damage Detector (T2D2) is a machine-based learning system that uses an image-analysis algorithm to search for and identify damage to building façades and roofs. The system was invented by Badri Hiriyur, director of Thornton Tomasetti’s CORE.AI lab. The T2D2 has been fed numerous photos from Thornton Tomasetti’s archives to teach it to identify problem areas. The company stresses that this technology is not meant to take the place of a qualified engineer conducting inspections, but rather to enable more frequent, automated inspections that might be able to identify possible problem spots and bring them to the attention of an inspector. It is hoped that this early intervention will help catch potential concern spots before they cause significant damage.
— ENR
