en.Wedoany.com Reported - On June 3, CNNC Huaxing officially released and put into use the "Nuclear Power Construction Safety Intelligent Monitoring Robot" integrated system at the Nanjing Binjiang Intelligent Construction R&D Base. The system consists of a fixed-point monitoring device, the "Nuclear Construction Safety Guardian Sentinel," and a mobile patrol device, the "Nuclear Construction Safety Patrol Robot Dog," promoting the transformation and upgrade of nuclear power construction safety management from manual monitoring to collaborative operations of "fixed-point duty + mobile patrol." Unlike previous single monitoring modes, the fixed-point device continuously monitors high-risk areas, while the mobile robot dog navigates complex environments to cover patrol blind spots. The two complement each other, achieving 24/7 monitoring without blind spots.
The "Nuclear Construction Safety Guardian Sentinel" is a portable fixed-point intelligent device weighing approximately 7 kg, which can be rapidly deployed in key work areas such as hot work zones and flammable material storage areas. The device integrates high-definition video, infrared thermal imaging, and multiple types of gas sensors, enabling 24-hour uninterrupted monitoring of hot work operations, fire hazards, dangerous gas leaks, as well as environmental temperature, humidity, oxygen, and carbon dioxide concentrations. It also supports facial authorization verification for workers, detection of labor protection attire, and identification of abnormal behaviors such as falls or phone use. All algorithms run locally on the edge device without relying on backend servers.
The "Nuclear Construction Safety Patrol Robot Dog" is a full-stack intelligent patrol robot specifically designed and developed for complex and hazardous areas that are difficult for personnel to access, such as pipe galleries, multi-story buildings, and confined spaces. It is equipped with LiDAR and visual sensors, supporting 3D high-precision mapping in indoor and outdoor complex environments, enabling centimeter-level autonomous navigation, and can automatically perform immediate, scheduled, and periodic patrol tasks.
Unlike fixed-point devices, the robot dog can actively patrol and dynamically capture hazards—using a thermal imaging dual-spectrum pan-tilt to monitor temperature anomalies in welding machines, gas cylinders, and work points, identifying fire risks caused by sparks and high-temperature debris in advance; simultaneously, it uses high-definition cameras to identify personnel violations. The backend "Safety Cockpit" centrally manages all edge devices, with alarm information automatically generating hazard rectification records, forming a closed-loop management system.
These two achievements represent another innovation from CNNC Huaxing's Cheng Xiaohua Innovation Studio, which has long been dedicated to intelligent nuclear power construction. Based on AI+ technology, the studio has achieved 13 industry-first results in areas such as robotic automatic welding, full-process automated processing of metal components, automated construction under complex on-site installation conditions, and digital management. The newly released "Sentinel + Robot Dog" integrated system further introduces AI vision, edge computing, and robot technology into safety supervision, enhancing the accuracy of identifying safety risks such as hot work operations, fire hazards, and dangerous gas leaks. It can significantly reduce manual patrol workload and improve intrinsic safety levels.
Currently, the system has been tested and validated at CNNC Huaxing's Nanjing Binjiang Base and the Bailong Nuclear Power site. In the future, it will be combined with aerial patrol drones to build a "air-ground-fixed point" three-dimensional nuclear power safety intelligent monitoring network, continuously driving the digital and intelligent transformation of nuclear power construction safety management.
This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com
