Active ultrasonic monitoring of reinforced and prestressed concrete structures has been investigated under laboratory conditions for over ten years. Methods originally developed for seismology, such as coda wave interferometry combined with embedded transducers, have proven highly sensitive to early-stage damage detection and robust against environmental factors such as temperature and humidity. Several real, or at least realistic, bridge structures have meanwhile been instrumented and monitored for months or even years. While being of utmost important in a time of ageing infrastructure these are not the only structures currently monitored, there are also example from underground, in particular a metro station and a nuclear waste disposal site.

Based on data acquired from these structures, some existing ones, some newly built, we can demonstrate that the measurement technology is stable and robust. Several experiments have proven that the technology can detect and characterize singular events, such as excess load or wire breaks in prestressing elements, as well as monitor subtle, slowly progressing material changes. The methodology is now starting to be used commercially.

Extensions to this technology are under development, such as using natural temperature changes to drive ultrasonic damage detection ('temperature modulation'), as well as other so-called nonlinear ultrasonic techniques. Proof-of-concept laboratory experiments have demonstrated the potential of these techniques.

Ultrasonic Monitoring is here to stay, but not on its own. As any other monitoring techniques, it has to be complemented with other techniques, such as its natural companion acoustic emission, fiber optic sensing and/or conventional sensors for temperature and strain.

Materials and structural components equipped with “sensing organs” and cognitive abilities may revolutionize many industrial sectors. Some of them, such as the civil engineering and aviation industries, would particularly benefit from the deployment of massively-dense intelligent sensor networks for structural integrity assessment. Although desirable, the realization of pervasive, permanently installed, real-time and large- or meso-scale monitoring solutions poses still major technical challenges, due to the severe energy consumption, weight and ease of installation requirements of these application fields. Moreover, the amount of new data generated may surpass the limit of maintenance systems to manage it, or even to gather it to central processing stations.

Such problems can be tackled by optimally distributing feature extraction and other intelligent tasks even beyond digital processors located on the extreme edge, i.e., directly to smart structural components, materials and sensors.

In particular, this talk will analyze: 1) the potential of novel sensing and pre-processing strategies based on recent research breakthroughs in nanocomposites and in functionalized (piezoelectric-, phononic-, resonant-, magneto- and electrostrictive-) metamaterials and sensing skins; 2) how these technologies can be coupled with electronic interfaces basing on Analog to Information techniques, and/or Mixed-Signal Computing with Non-Volatile Memories, and/or Approximate Computing for an omni-layer optimization of the operational bandwidth, autonomy, robustness, and efficiency of the SHM systems.