China Telecom, ZTE and AGIBOT Unveil 5G-A-Powered Embodied Intelligence Solution for Hazard Detection in Industrial Parks

LongbridgeAII'm LongbridgeAI, I can summarize articles.

China Telecom, ZTE Corporation and AGIBOT (Zhiyuan Robotics) have jointly rolled out a 5G-A-powered embodied intelligence inspection solution for chemical industrial parks, with its pilot deployment launched at the Shangyu Chemical Industrial Park in Zhejiang Province. The solution integrates AI-native networks, edge computing and embodied intelligence to build a unified device-edge-cloud collaborative system, addressing long-standing pain points in chemical safety inspection.

As one of the top 10 chemical industrial parks in China, Shangyu Chemical Industrial Park spans 175 square kilometers with an annual output value of over USD 40 billion. It is the world's largest dye production base and Asia's largest peroxide manufacturing hub. The park hosts a high density of high-temperature, high-pressure and highly corrosive production environments, rendering safety inspections considerably more challenging and hazardous than those at ordinary manufacturing facilities.

The traditional manual inspection model is constrained by the park's vast scale, heavy workload and hazardous environment, exposing frontline workers to severe occupational health risks. Compounded by the high professional threshold for the role, the shortage of qualified inspection personnel continues to worsen, emerging as a core pain point for the park's safe and efficient operation.

To address this industry-wide challenge, a collaborative inspection solution combining human inspectors and robot dogs is deployed. The robot dogs carry out routine inspections across most of the plant, while professional inspectors perform detailed checks in areas inaccessible to robots. Inspectors can also remotely operate and monitor the robot dogs from a central control room, maximizing both operational safety and inspection precision while improving inspection coverage efficiency.

Nevertheless, the complex plant environment creates two core bottlenecks for robotic inspections. First, dense pipe corridors and equipment cause severe Wi-Fi signal attenuation and widespread blind spots, which cannot support stable, reliable communication links. Second, high-risk areas require 24/7 monitoring, and the concurrent operation of multiple robots places extremely high demands on computing power. Conventional device-side computing suffers from high deployment costs and low utilization of distributed resources, making it unable to meet the requirements of high-concurrency, uninterrupted inspections.

To overcome these barriers, the solution delivers an end-to-end intelligent inspection system underpinned by device-edge-cloud synergy. The device side retains only lightweight computing power for basic perception and emergency response actions. The edge side handles compute-intensive tasks including scene understanding and global path planning. Powered by 5G-A and an all-in-one compute-network platform, the solution ensures industrial data security, delivers air interface latency below 10 ms and enables unified scheduling of communication and edge computing resources. The cloud side handles complex cross-scenario decision-making and multi-robot dogs formation coordination to support concurrent multi-task inspection coverage across the park's complex operating conditions.

To date, the pilot has deployed 26 robot dogs across 5 key chemical enterprises in Shangyu, covering 12 types of high-risk inspection scenarios including spray tower workshops and pipe corridor zones. The deployment has drastically reduced workers' exposure to hazardous environments, accelerated hazard warning response, and effectively lowered enterprises' operating costs. Having been fully validated its adaptability to chemical scenarios, the solution can be rapidly scaled to other high-risk industries such as electric power, mining, metallurgy and oil and gas, delivering broader safety and efficiency benefits.

Contributed by ZTE.

Source: The Register The copyright of this article belongs to the original author/organization.

The views expressed herein are solely those of the author and do not reflect the stance of the platform. The content is intended for investment reference purposes only and shall not be considered as investment advice. Please contact us if you have any questions or suggestions regarding the content services provided by the platform.