Cognitive Systems & Robotics Laboratory

Innovative Surveillance System for Timely Defect Prediction in Critical Infrastructures ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎ ‎

Innovative Surveillance System for Timely Defect Prediction in Critical Infrastructures

Aggravation of extreme climate phenomena in recent years in Europe as well as worldwide is an undeniable fact that underlines the importance of meteorological prediction and early warning infrastructure. Consequently, the continuous and orderly function of remote and mobile weather stations is paramount and needs to be ensured at any time. Thus, the requirement for robust remote weather infrastructure indicates the importance of fault monitoring and identification technologies for remote infrastructure. In this direction, EARLY aims to design and develop a novel and comprehensive system for multimodal fault monitoring, identification, and prediction, focusing on application to remote infrastructure and specifically to remote weather stations and weather radar infrastructure. The envisioned system will be capable of acquiring multi-sensory data relevant to infrastructure condition through locally embedded sensors as well as Unmanned Aerial Vehicles (UAVs) and ground robots that will be equipped with multi-sensory observation platforms. In addition, EARLY will enable data analysis using sensor fusion and deep-learning methods, with the aim of fault identification and prediction, focusing on embedded edge processing. At the same time, a dynamic and autonomous robot duty assignment and observation plan adaptation will occur based on the robot’s condition and analysis of acquired infrastructure function data. Last, a central decision support system will be developed that will be responsible for the acquisition, mapping and presentation of semantic information concerning potential infrastructure faults or malfunction, which is expected to significantly contribute to maintenance-related decision making. Through the envisioned predictive monitoring and fault identification, EARLY aims to minimize the cost of monitoring of remote weather infrastructure, improve the distribution of skilled workforce through minimization of manual inspection, and increase the effectiveness of maintenance through real-time decision support. Successful realization of EARLY goals is expected to offer strategic improvements to the ability to predict and prevent faults and malfunction in remote infrastructure, contributing thus to the orderly function of monitored infrastructure. Ultimately, the technological developments stemming from successful completion of EARLY are expected to offer a key advantage to bodies overseeing remote weather infrastructure and decisively contribute to the continuous and accurate forecast of extreme weather events, in turn contributing significantly to civil safety and environmental protection.