ErfASst

The road network is an important element of social life, prosperity and the supply of goods. In Europe, over 90 % of inland traffic is handled via terrestrial transport routes. Engineering structures in the form of bridges are neuralgic, singular objects that are subject to considerable strain due to the increase in heavy goods traffic and whose failure can have enormous consequences for the functionality of the transport network. Failures in this network can have far-reaching consequences for life and limb as well as for the security of supply. The condition assessment and effective maintenance of these bridges currently represent a major challenge, as the condition assessment and evaluation is based on many small-scale, manual and non-automated assessment processes and evaluation steps. This is where ErfASst comes in.

A semi-automated condition assessment system was developed as part of the ErfASst project. The system consists of a stationary or mobile optical camera system that photographs the bridge surface to be inspected, a crack detection tool that uses machine learning to automatically recognise cracks in the images and an assessment tool that links the detected crack data with a condition score. In addition, the progress of corrosion (chloride-induced corrosion) of the concrete reinforcement can be estimated by combining it with other data streams. In this way, the progression of damage in the structure can be mapped and predicted so that new buildings or repair measures can be planned more efficiently, resulting in fewer unscheduled disruptions and thus an increase in resilience.