ASINVOS

Most public infrastructures in urban areas are aging and need frequent inspection. Distress detection and solid management for maintenance is the key to guarantee their permanent availability. Current evaluation methods are mostly focused on visual inspection which is done manually and therefore requires excessive manual labor. Thus, the time between the actual inspection and the final evaluation may take up to several months. In the meantime, small damages like cracks can lead to substantial downtimes with a high impact for the population.

The ASINVOS project aims to automate this process to a high degree by applying machine learning techniques. The basic idea is to train a self-learning system with manually annotated data from previous inspections so that the system learns to recognize underlying patterns of distress. Once the system can robustly identify intact infrastructure, it can reduce the human amount of work by presenting only distress candidates to the operator. This helps to speed up the inspection process significantly and simultaneously reduces costs. Furthermore, inspection intervals can be reduced which helps to remedy deficiencies in time.

The system is applied to two types of infrastructure, namely water pipes, and public roads, to show the versatility of the self-learning system.