In response to the current issues of inadequate community instrument inspection, difficulty in manual inspection operations, and the lack of effective solutions for high-risk areas, our team proposes a "Community Safety Inspection System based on the Internet of Things (IoT)". This system consists of three major components: hardware modules, software services, and an interaction system. It mainly includes five core components: sensor modules, quadruped robots, cloud-based databases, front-end visualization platforms, and back-end support systems. The goal is to address the industry's demand for efficient and cost-effective community safety inspections.
Firstly, the system collects data indicators such as temperature, humidity, or content levels from the field through sensor modules. This data is then transmitted in JSON format to a cloud database for storage and visualization on a web-based dashboard, allowing personnel to monitor the data. Secondly, the system utilizes quadruped robots based on machine vision algorithms to extract instrument readings from video streams and provides feedback for manual online assessment, enabling diagnostics of instrument status. Ultimately, this achieves an intelligent community inspection workflow of "sensor module monitoring - robot pre-inspection - manual online assessment - instrument issue diagnosis".