Reinforcement learning-based IoT sensor scheduling strategy for bridge structure health monitoring

Internet of Things (IoT) based Bridge Structural Health Monitoring (BSHM) is a hot topic in the field of civil engineering and computer science, and has been widely concerned by academia and industry. The lifetime of the sensors is much less than that of the bridge, which is one of the main technical bottlenecks in BSHM. Therefore, how to effectively improve the network lifetime is the focus of current research. Based on reinforcement learning and Fisher information matrix, this paper proposed a node sleep scheduling strategy by using the learning automata model and confident information coverage (CIC) model to learn the optimal sensor sleep scheduling strategy through cooperative sensing among nodes. Fisher information matrix, which is widely used in civil engineering, was introduced to define the node sleep scheduling problem as a multi-objective optimization problem. With information validity as the modal assurance criterion, the network performance was measured by combining energy efficiency, network coverage requirements and network connectivity. While ensuring the network connectivity and coverage requirements, the system parameter identification error is minimized and the network life is maximized. Through the simulation of jiangbei Bridge in Guangdong, China, the effectiveness, energy efficiency and applicability of the proposed scheme are verified.