Cracksight applies AI-powered computer vision to identify and classify surface anomalies, cracks, fractures, material spalling, corrosion pockets, and weathering patterns across heritage fabric. The system analyses drone imagery, photogrammetric scans, and optical data to generate damage maps overlaid on 3D models. Early detection enables targeted interventions before structural compromise.

DYNAMO revolutionises heritage inspections using autonomous drones equipped with SLAM (Simultaneous Localisation and Mapping), NeRF (Neural Radiance Fields), and SfM (Structure from Motion) to access impossible-to-reach areas like bell tower interiors and confined wooden structures. Drones capture high-resolution 3D data, which deep learning algorithms analyse to recognise damage patterns, material degradation, and mechanical wear. Data feeds directly into finite element (FE) models to simulate structural behaviour and detect faults. Key targets: 90% accuracy in identifying structural damage and 70% reduction in hazardous on-site human intervention.

MUSICA harnesses cosmic-ray muons (naturally occurring particles) to non-invasively map internal density variations in heritage structures. Revealing voids, decay zones, reinforcement bars, and wooden beam degradation that external inspection cannot detect. Unlike X-rays or neutrons, muography is safe and portable, ideal for immovable monuments. Detector networks strategically deploy around/within heritage objects, transmitting muon flux data to cloud-based processing for 3D tomographic reconstruction. Machine learning algorithms optimise detector placement and reconstruction, whilst Monte Carlo simulations support algorithm training. This results in internal structure maps that are integrated into dynamic digital twins and AR/VR visualisations for collaborative conservation planning.