Development of IoE-Edge-AI Architecture For Landslide Early Warning System

Landslides are a prominent natural disaster in India that causes huge socioeconomic losses yearly. The advent of technologies like Internet-of-Everything (IoE), Wireless Sensor Network (WSN), Artificial Intelligence (AI), and big data has been proven invaluable in the field of disaster monitoring. The proposed research develops a novel system to collect real-time data using IoE-Edge-AI architecture to mitigate the severity of landslides. IoE is based on all-around connectivity, intelligence, and cognition that integrate people, processes, data, and objects in the common network to generate valuable information for real-time decision-making. The architecture enables an intelligent environment (edge servers) in the vicinity of the developed IoE-end node. On the one hand, the IoE-end node is embedded with multiple sensors to sense site-specific parameters. On the other hand, Intelligent Edge provides an interface for local residents to share live site information (text, image, or video) to participate in real-time decision-making. The suggested method not only empowers community involvement in the landslide early warning system but also overcomes the existing issues of connection loss between nodes and remote cloud as well as nodes and edge by proving an alternative edge in the multi-edge architecture. This makes the system reliable and fault-tolerant during harsh geological environments. The intelligent edge server collects data from IoE-end nodes and the community and processes and analyses it to generate an optimum threshold for the Landslide Early Warning System (LEWS).