DAWN

CFMS’ Robotics and Autonomous Systems experts bring their technical knowledge within machine learning (ML) and artificial intelligence (AI), combined with its established data centre to play an active role in the design and development of a pipeline dedicated to the detection, classification and segmentation of visual data for two specific inspections of the topskin panel of a wing at different assembly stages.

CFMS inspects both the mounted brackets and the drilled holes and fastener installation, aiming to assess the feasibility of utilising AI systems to enhance the speed and quality of inspections, traditionally performed manually. Ultimately, this exploration may lead to the automation of certain tasks, reducing the time and cost of manufacturing aircraft wings.

To analyse this information, our RAS team utilises advanced technologies for more effective data processing, leveraging cutting-edge ML tools for image analysis. Given the limited availability of training data, we take advantage of CAD models and photo-realistic simulators to generate a large volume of synthetic images, reaching hundreds of thousands, these technologies can enhance the speed and quality of data analysis. This approach proves to be highly cost-efficient due to the utilisation of our on-site data centre.