On April 10, Zhongke Zhiyun Technology Co., Ltd. officially released SIEA – CORE (Super Industrial Equipment Agent).
This is a global intelligent agent for industrial equipment built with the self-developed World Model for Industrial AI as the core. It enables industrial equipment to have the core ability to understand the physical world and predict dynamic processes, and helps traditional industrial equipment achieve the transition from ‘human operation’ to ‘autonomous intelligence’.
This move is not only in line with the country’s future industrial development orientation, but also injects strong hard-core AI momentum into cultivating new productive forces and promoting industrial upgrading.
With the in-depth implementation of the Industry 4.0 strategy and the widespread application of new technologies such as artificial intelligence, big data, and cloud computing, China’s intelligent manufacturing equipment market is facing unprecedented development opportunities.
According to predictions by many institutions, by 2030, China’s intelligent manufacturing equipment market is expected to exceed 4.5 trillion yuan, with a compound annual growth rate between 12% and 16%.
At the same time, the country vigorously promotes large-scale equipment updates in the industrial field and leads the transformation of manufacturing into high-end, intelligent, and green.
Zhongke Zhiyun’s self-developed industrial world model aims at the pain points of traditional industrial automation’s difficulty in single-scenario customization and multi-scenario adaptation. It achieves technological breakthroughs by internalizing the underlying physical laws and completes the upgrade from ‘scenario-specific customization’ to ‘all-scenario, all-purpose, all-autonomous’, innovating the underlying logic of industrial intelligence and breaking industrial control technology barriers.
Different from other embodied intelligence companies, Zhongke Zhiyun focuses on the field of industrial equipment AI with real scenarios, real applications, and real income.
Closely following policy requirements, Zhongke Zhiyun has in-depth layout of the two core industry tracks of future information and future space.
The SIEA-CORE released this time can be widely adapted to various industrial scenarios. The first batch has been applied to key equipment such as tower cranes, bridge cranes, stackers and reclaimers, ship loading and unloading machines, covering important fields such as construction, ports, energy, steel, chemical industry, etc.
It is not only suitable for new factories, but also enables the intelligent upgrade of existing equipment, setting a benchmark for the intelligent upgrade and digital transformation of traditional industrial equipment.
SIEA-CORE is like an ‘intelligent driving system for industrial equipment’.
Through ‘multi-dimensional perception’, it can ‘understand’ the surrounding environment like a human being, predicting and preventing possible future collisions; and through ‘virtual training + real application’ technology, it can complete thousands of operational exercises in the digital world, thereby completing tasks accurately and safely in real scenarios.
The current intelligence of industrial equipment is equivalent to the L2 stage of intelligent driving, and the birth of SIEA-CORE is expected to push the industry towards the L3-L4 level of intelligence.
The core design concept is to take the physical world as the starting point to create industrial autonomous intelligence that understands mechanics, the environment, and needs.
Different from the traditional large language model, the technical base of SIEA-CORE is the industrial world model that Zhongke Zhiyun continuously iterates. By learning all the data of industrial equipment in real operations and simulated scenarios, it can more accurately grasp the physical movement laws of the equipment, realize understanding of the physical world, predict dynamic processes, comprehensively perceive various physical elements and equipment operating status in the operating environment, deduce key processes such as equipment movement trajectories and force changes in advance, so that industrial equipment has autonomous cognitive and forward-looking decision-making capabilities, and supports intelligent management and control of the entire operation process.
The fusion of perception and semantic SLAM technology (semantic simultaneous positioning and map construction) forms an important support for the implementation of SIEA-CORE scenarios.
While the multi-sensor fusion perception system collects global environmental data in real time, semantic SLAM technology simultaneously completes scene modeling and operating status analysis, providing the “brain” with accurate scene recognition, assisting it in identifying operational risks in advance and issuing safe and effective control instructions.
Sim2Real technology (simulation-to-reality migration technology) solves the difficulty and high cost of real data collection in industrial scenarios. It improves model efficiency through full-condition simulation training, significantly shortens the implementation cycle, and accelerates the large-scale application of cutting-edge AI technology in the industrial field.
In terms of scene verification, taking tower cranes as an example, their safety control has always been a pain point in the industry.
Since 2018, Zhongke Zhiyun has cooperated with domestic leading industrial manufacturing companies to carry out AI intelligent technology research and development and engineering implementation around the safety challenges of tower crane operating scenarios.
The intelligent safety center built by SIEA-CORE for this type of industrial equipment can avoid human operation risks from the source and accurately adapt to various complex operating conditions.
Actual measurement data shows that at the large construction sites of China State Construction Engineering Corporation, the operating efficiency of tower cranes equipped with SIEA-CORE can reach 85% of that of experienced skilled workers, and the all-weather working efficiency is 10% higher than that of skilled workers. It also supports three-axis linkage and five-speed operation, achieving dual improvements in safety and efficiency.
In addition, Zhongke Zhiyun has deepened cooperation with China Communications Information Technology Group to carry out intelligent upgrades for stackers and reclaimers, ship loading and unloading machines and other equipment, achieving 24/7 unmanned intervention and increasing loading and unloading efficiency by 15%.
Dr. Li Yuan, chief technology officer of Zhongke Zhiyun, said: ‘The core competitiveness of future industries comes from the deep integration of cutting-edge technology and real industrial scenarios.
Zhongke Zhiyun has been deeply involved in industrial scenarios for a long time and has accumulated a large amount of high-quality real-scene application data.
This release marks that Zhongke Zhiyun has officially entered the main track of “intelligent industrial equipment”. We will continue to delve into construction, ports, energy, steel, chemical and other scenarios, promote more industrial equipment to achieve unmanned and intelligent upgrades, and provide replicable and promoteable “Zhongke Zhiyun solutions” for my country’s new industrialization construction, high-quality development of manufacturing and future industrial innovation breakthroughs.
’
