In 2021 in Europe more than 100k vessels will be inspected through confined space entry in the PetroChemical Industry. This is done to look for signs of cracking, corrosion or other damage, to accurately assess that damage and, if necessary, to effect repairs. To avoid confined space entry by humans, a stated safety goal of the European Chemical Industry, robots will need to take over these tasks. Currently, robotics still rely on manual interpretation of the data which requires significant manual effort (time), is subjective and ties up highly skilled resources. Also, when damage indications are found, human entry is generally necessary to validate the findings and determine the exact dimensions and nature of the defect and the need for repair.
Real-time robotic AI supported interpretation of visual inspection with follow up dimensioning can overcome these disadvantages and is believed to transform robotic inspections and unlock huge productivity gains for all stakeholders. In this project, a unique hybrid vacuum crawler robot equipped with ultra-HD cameras and a laser 2D/3D scanner enters the vessel which requires internal inspection. The XaaSLabs real-time AI detection algorithm monitors the inspection video cameras during the inspection process. If a potential defect is determined to be of sufficient interest, the 2D/3D scanner will be used to accurately characterize and quantify the nature and severity of the defect.