Congratulations to Prof Fu Yuguang on the award of MOE AcRF Tier 1

Heading towards Smart Nation, Singapore is establishing pervasive digitalization to improve the quality of people’s life. As one of the key domains for digitalization, civil infrastructure desires a great number of sensors for various purposes, such as construction management and in-service inspection, which imposes challenges of data inundation and sensor malfunction. This calls for novel techniques to preserve data quality and achieve sensing autonomy for long-term deployed monitoring systems. Conventional studies handle it as part of big data challenges, where data cleansing and mining are conducted after it is collected in a central station. However, faulty sensors still generate/transmit misleading data while consuming power and memory continuously. 

This study will provide an effective artificial intelligence (AI)-based solution embedded in IoT sensors to minimize resource consumption and maximize effective information for autonomous long-term monitoring of civil infrastructure. To achieve this, we will develop an AI framework where big data analytics, edge computing, and structural health monitoring meet. We propose two key techniques: (i) smart adaptive triggering schemes to support on-demand sensing while preserving energy and memory efficiency; and (ii) self-diagnosis strategies to detect, identify, and report faults in IoT sensors while maintaining high-quality data acquisition. The proposed solution will help to increase the efficiency and reduce maintenance costs for civil infrastructure monitoring systems in Singapore and offer the potential to boost pervasive digitalization for Smart Nation in Singapore.