根据MIR Databank最新发布的统计数据，2025年上半年，中国工业机器人市场出货量排名中，南京埃斯顿自动 化股份有限公司（以下简称"埃斯顿"）连续两个季度位居首位，市场份额达到10.5%（含旗下品牌埃斯顿酷 卓），成为国内出货量最大的工业机器人供应商。同期，中国工业机器人市场国产化率进一步提升至55.3%，较 2024年全年提升约5个百分点。 作为全球工业机器人竞争最激烈的区域市场，中国汇聚了超过95%的国际主流机器人品牌。在此背景下，本土企 业的持续增长引发业界关注。分析认为，埃斯顿近年来在技术平台升级、场景工艺深耕及全球化布局方面的系统 性投入，是其市场份额稳步提升的重要动因。 技术底层重构：从"编程设备"向"工业智能体"演进 2025年初，埃斯顿推出新一代开放式控制平台NGC，并在德国Automatica展会上发布支持二次开发的ERI （Estun Robot Interface）操作平台。该平台提供多类API接口、开发套件及SDK工具，支持4ms级实时通信，可 实现多机器人协同、外部轴联动以及与视觉、力控等传感器的深度集成。 业内专家指出，此类平台的开放性有助于加速机器人在复杂产线中的部署效率 ...

Core Insights - The article discusses the transition from automation to autonomy in manufacturing, driven by advancements in artificial intelligence, enabling systems to make independent decisions and adapt dynamically to complex industrial environments [1][2][3] - The concept of "industrial agents" is introduced, which possess high autonomy, allowing them to perceive, plan, and execute tasks without human intervention [2][3] - The report highlights the current state and future trends of industrial agents based on a collaboration between Siemens and Zhiding Technology, along with interviews from various manufacturing enterprises and industry experts [1][6] Industrial Agents Overview - Industrial agents represent a significant evolution in AI applications, moving from basic automation to advanced decision-making capabilities [2][3] - Key characteristics of effective industrial agents include knowledge, understanding, reasoning, and execution [3] - The transition to industrial agents is seen as a gradual revolution, with companies already implementing AI technologies for decision support in supply chain management and quality control [3][4] Application Scenarios - The report identifies three core application areas for industrial agents: production manufacturing (44%), research and design (32%), and operation maintenance (25%) [8] - In production, industrial agents enhance efficiency and quality by enabling dynamic scheduling and real-time parameter optimization [9][10] - The chemical and automotive sectors are actively exploring the use of industrial agents, with some companies achieving full process integration in their operations [7][9] Challenges and Adoption - Despite the potential benefits, many companies remain cautious, with 43% of surveyed manufacturers not yet deploying industrial agents [6] - The primary challenges include high deployment costs (63% of companies cite this as a major concern) and a lack of skilled personnel who understand both production and AI [13][14] - The technology is still considered immature by 40% of companies, highlighting the need for industry-specific models and reliable data [16] Future Trends - The future of manufacturing is envisioned as a network of interconnected industrial agents capable of real-time decision-making and flexible production [19][20] - Companies are increasingly open to collaborating with external technology providers to co-create solutions, indicating a shift towards a more integrated industrial ecosystem [20] - The article concludes that the evolution of industrial agents could redefine manufacturing competitiveness, moving from a reliance on labor to leveraging technological and intelligent advantages [20]

Core Insights - The global industrial intelligence market is expected to exceed 3.5 trillion yuan this year, with China holding over 40% of the market share [1] - The Ministry of Industry and Information Technology emphasizes the development of industrial intelligence systems to enhance AI applications in the industry, focusing on typical scenarios for research and application [1] Industry Overview - Industrial intelligence systems integrate AI, big data, and cloud computing, enabling intelligent decision-making and collaboration, significantly transforming manufacturing processes [2] - These systems reduce reliance on individual experience in R&D, shorten development cycles, and optimize production scheduling and equipment maintenance [2] Market Trends - Various regions in China are implementing policies and initiatives to accelerate the application of industrial intelligence systems, moving from laboratory settings to practical industry applications [3] - The Yangtze River Delta electronic information industry cluster has improved cross-enterprise process design efficiency and supply chain responsiveness through collaborative platforms [3] - By 2028, AI spending by Chinese industrial enterprises is projected to reach 90 billion yuan [3] Challenges and Solutions - Effective application of industrial intelligence systems requires overcoming data bottlenecks, such as data silos and complexity, necessitating the integration of various data sources [4] - Companies need to establish robust data management systems to support advanced intelligent applications [4] Technical Development - The complexity of industrial scenarios demands the development of adaptable and flexible AI platforms that can address technical challenges like computing power adaptation and model compression [4] - There is a need for deep integration of technological advancements with industry-specific demands to maximize the value of industrial intelligence systems [4] Talent Development - The successful implementation of industrial intelligence systems relies on a skilled workforce that understands both industrial mechanisms and AI technologies [4] - Emphasis on talent cultivation and education in AI across all levels is crucial for enhancing the overall AI competency within the industry [4]

Core Insights - The challenges of achieving "AI + industrialization" include data acquisition and governance, as highlighted by Siemens at the 25th China International Industry Fair [1][4] - Siemens has signed agreements with nearly 40 companies to expand its local ecosystem and has showcased its first pilot application of the Industrial Copilot based on generative AI in China [1][3] Group 1: Industry Trends - Over 80% of enterprises recognize the value of industrial AI in enhancing production efficiency and reducing operational costs, yet 43% have not deployed industrial AI, and only 8% have achieved multi-scenario applications [3] - Key barriers to deploying AI industrial systems include high deployment costs, a lack of skilled professionals, and insufficient technological maturity [3] Group 2: Company Developments - Siemens has developed a new generation of smart assembly equipment for electric vehicles in collaboration with Zhongke Motong, marking the first pilot application of its generative industrial AI assistant in the Chinese market [4] - The new industrial PC product launched by Siemens is tailored for the Chinese market and compatible with the local operating system Kirin, demonstrating a proactive response to local demands [5] Group 3: Financial Performance - Siemens reported a 28% year-on-year increase in new orders, reaching €24.7 billion, and a 5% revenue growth to €19.4 billion for the third quarter of fiscal year 2025, although the software business within the digital industrial group faced challenges compared to the previous year [6]

Core Insights - The article discusses the transformation of the industrial internet, emphasizing the shift towards "Industrial Internet 2.0" with a focus on smart factories and intelligent platforms as core strategies for enhancing efficiency and decision-making in manufacturing [1][2]. Group 1: Industrial Internet Transformation - The industrial internet is evolving, with a core path defined as smart factories combined with intelligent platforms, aiming to address challenges such as rising costs and intensified competition in the manufacturing sector [1][2]. - The concept of "Industrial Intelligence" is introduced as a new paradigm for scaling Chinese manufacturing, highlighting the importance of intelligent decision-making and autonomous optimization alongside connectivity and data [2][3]. Group 2: Technological Implementation - The company has developed a low-code platform called "Lingxi Industrial Operating System," which streamlines the development process, resulting in a 59% increase in development efficiency and a 70% reduction in operational costs [2][3]. - Six key scenarios have been identified for practical application, including industrial governance and supply chain management, which have been validated through extensive real-world testing [2][3]. Group 3: Global Expansion Strategy - The company is preparing for international expansion, aiming to export its industrial internet and intelligent capabilities, particularly targeting the light industry sectors that require flexible production lines and rapid product iteration [4]. - The shift from exporting products to exporting capabilities signifies a strategic transformation, focusing on providing integrated smart solutions rather than just hardware [4]. Group 4: Talent Development - A new talent cultivation philosophy called "PhD in the Workshop" is being implemented to bridge the gap between technology and manufacturing processes, fostering a new generation of professionals who understand both domains [5]. - Collaborations with universities to establish industrial internet academies aim to produce versatile talents capable of engaging in real projects and gaining practical experience [5].

Core Insights - The "Artificial Intelligence Technology Application Reference Guide for the Manufacturing Sector in Jiangsu Province (2025 Edition)" has been released, marking the first national guideline focused on AI applications in intelligent manufacturing scenarios [1] - The guide is developed by the East China Branch of the China Electronic Technology Standardization Institute, leveraging its expertise in standard formulation and evaluation for intelligent manufacturing [1] - The guide outlines a systematic approach to AI applications in manufacturing, categorized into four implementation stages: "Entry Level," "Basic Level," "Advanced Level," and "Forward-looking Level," covering the entire lifecycle of manufacturing from single-point applications to system-wide optimization [1][2] Implementation Framework - The "Entry Level" section focuses on initial AI applications, proposing 9 typical scenarios that utilize perception and recognition technologies to assist human operations in areas such as warehouse management and quality inspection, thereby strengthening the digital foundation of enterprises [2] - The "Basic Level" targets enterprises with some automation capabilities, presenting 20 scenarios aimed at enhancing cross-link efficiency in factory planning and supply chain collaboration through technologies like digital twins and knowledge graphs [2] - The "Advanced Level" is designed for leading enterprises, offering 31 scenarios that promote intelligent collaboration and autonomous decision-making across all processes, aiming for self-evolving and self-optimizing production systems [2] - The "Forward-looking Level" introduces the concept of "Industrial Intelligence," exploring future manufacturing systems with autonomous perception, analysis, decision-making, and execution capabilities, facilitating the transition from "intelligent" to "autonomous" manufacturing [2] Future Directions - The East China Branch of the China Electronic Technology Standardization Institute plans to accelerate the research and construction of an "AI Factory" evaluation system, aiming to establish scalable technical pathways and standards to support the intelligent transformation of manufacturing across the province and the nation [2]

Core Insights - The release of AI technologies, particularly the DeepSeek R1 model, has significantly lowered the cost of deep thinking models, leading to widespread adoption of AI across various sectors [1] - The Chinese government has accelerated the penetration of AI into multiple industries through the "Artificial Intelligence+" initiative [1] - The industrial AI market in China is projected to exceed 120 billion yuan by 2025, with a compound annual growth rate of 28.6% [1] Industrial AI Development - The consensus in the industrial sector has shifted from questioning the integration of AI to increasing investments in AI technologies [3] - Major companies have launched various industrial AI models, including those from iFLYTEK, China Unicom, and China Mobile, indicating a competitive landscape [3] - The TPT model developed by Zhongkong Technology has demonstrated significant efficiency improvements in industrial processes, such as reducing pH adjustment time by over 80% [3][4] Industrial AI Agents - Industrial AI Agents are emerging as the optimal form for implementing AI in industrial settings, characterized by autonomous perception, cognition, decision-making, and learning capabilities [4] - The global industrial intelligence market is expected to surpass 3.5 trillion yuan, with China holding over 40% of the market share [4] - The TPT 2 model represents a paradigm shift, enabling autonomous decision-making and optimizing production processes [9][10] Market Response and Adoption - Zhongkong Technology has established a strong presence in the industrial AI sector, serving over 37,000 clients and deploying more than 100,000 industrial control systems [5] - The TPT model has received significant market recognition, with 518 pre-orders and 112 confirmed transactions following its launch [16] - The establishment of the "Industrial AI Data Alliance" aims to create a collaborative ecosystem for data sharing and security, enhancing the value of industrial data [18] Financial Performance - Zhongkong Technology reported a revenue of 9.139 billion yuan in 2024, with a year-on-year growth of 6.02%, and a notable increase in industrial software revenue by 20.7% [19] - The gross margin for industrial software has significantly improved, reaching 54.03% in the first half of 2025, up from 31.67% in the previous year [19] - The company is positioned as a leader in the industrial AI sector, with its platformization strategy driving growth and innovation [21]

Core Insights - The emergence of industrial intelligent agents represents a significant transformation in China's manufacturing sector, driven by advancements in artificial intelligence (AI) [1][4] - Midea Group's washing machine factory in Jingzhou has been recognized as the world's first multi-scenario intelligent factory, marking a milestone in the industry [1][4] Group 1: Intelligent Factory Concept - The intelligent factory is likened to a human, where the "brain" manages task scheduling, the intelligent agents act as a neural network, and smart terminals execute tasks [2] - Unlike traditional smart factories focused on automation and digitization, intelligent factories emphasize high autonomy and continuous evolution, enabling proactive perception and decision-making [2][4] Group 2: Technological Innovations - AI glasses in the factory enhance inspection efficiency, reducing the first inspection time from 15 minutes to 30 seconds [3] - Humanoid robots, such as "Mei Luo," are capable of performing multiple tasks across functions, breaking traditional job barriers and enhancing flexibility in manufacturing [3][4] Group 3: Performance Improvements - Midea Group has deployed 14 intelligent agents covering 38 core production scenarios, achieving over 80% efficiency improvement and a 90% increase in production scheduling response speed [4] - The company has committed to significant R&D investments, totaling 400 billion yuan over the past three years and planning an additional 500 billion yuan in the next three years to advance AI and robotics [4] Group 4: Policy Support - The Chinese government has issued policies to promote AI integration in manufacturing, emphasizing the need for intelligent collaboration across all industrial processes [5] Group 5: Future Goals - Midea Group aims to create a self-regulating intelligent ecosystem that can adapt and evolve, ultimately providing a reusable intelligent factory model for global manufacturing [6] - The company plans to replicate the intelligent factory solutions across its global operations to enhance China's competitive position in the manufacturing sector [6]

Core Insights - The industrial intelligent agent is a fusion of large models, industrial mechanisms, and machine learning, generating significant economic value and driving innovation in industrial applications [1][2] - The global industrial intelligence market is expected to exceed 3.5 trillion yuan this year, with China accounting for over 40% of the market share, indicating a rapid acceleration towards the era of industrial intelligent agents [1] Group 1: Definition and Functionality - Industrial intelligent agents are designed specifically for industrial production, possessing autonomous perception, cognition, decision-making, and learning capabilities, distinguishing them from general intelligent agents [2] - These agents can understand high-level and natural language commands, transforming human-machine interaction by allowing direct command execution without manual software operation [2][4] - The integration of multi-modal perception, large model task planning, and refined motion control enhances the autonomous operational capabilities of robots in complex industrial environments [3] Group 2: Applications and Benefits - Industrial intelligent agents are shifting R&D from experience-based trial and error to an intelligent-driven paradigm, significantly reducing R&D cycles and enhancing design combinations [3] - In manufacturing, these agents facilitate the transition from automation to autonomy, optimizing production scheduling, equipment maintenance, and cross-system collaboration [3][5] - The implementation of intelligent agents has led to a 60% reduction in process preparation time and a 20% increase in order fulfillment rates, showcasing their efficiency-enhancing capabilities [6] Group 3: Challenges and Future Directions - The deployment of industrial intelligent agents faces challenges such as technology maturity, data isolation, and the complexity of industrial environments, which affect adaptability and reliability [7] - Safety concerns are paramount, as intelligent agents operate through autonomous code generation, exposing them to potential security threats like API vulnerabilities and code supply chain issues [7] - Strengthening infrastructure, establishing standard systems, and creating experimental ecosystems are essential for the effective deployment and integration of industrial intelligent agents [8]

Core Viewpoint - Estun has maintained its position as the leading player in China's industrial robot market for two consecutive quarters, achieving a market share of 10.5% by the first half of 2025, which has contributed to an increase in the domestic production rate of industrial robots to 55.3% [1][3]. Group 1: Core Competitiveness Upgrade - The integration of AI and robotics has led to the development of industrial intelligent bodies, enhancing usability and deployment ease through solutions like teach-free welding and visual-guided random grabbing [5]. - Estun's NGC control platform and ERI implementation control platform support high openness for secondary development, facilitating multi-robot collaboration and real-time data processing [5]. - The evolution of intelligent robots is transforming ordinary workers into skilled technicians, particularly in low-density application scenarios, thereby expanding the market size for robots [5]. Group 2: Scene-based Process Integration - The incorporation of AI technology has broadened the application scenarios for robots, while advancements in scene-based software are crucial for deepening these applications [6]. - Estun's certification of its explosion-proof spraying robot in July 2025 demonstrates its capability to cover various processes in metal processing, establishing a strong foundation for entering the automotive sector [6]. - The strategy of combining horizontal processes with vertical industries allows Estun to offer modular solutions tailored to customer needs, marking a significant shift from single machine sales to comprehensive solutions [6]. Group 3: Equal Opportunities for Domestic and Foreign Brands - China's robot industry has emerged as a core player in the global market, enhancing competitiveness and ensuring the safety of the domestic industrial chain [7]. - By the end of 2024, Estun will have established 75 business outlets globally, with the completion of its Poland factory, creating a supply chain structure that includes three overseas and five domestic production bases [7]. - The shift from following to surpassing foreign brands reflects the robust competitiveness of China's robot industry [7]. Group 4: Market Transformation - The transformation in the Chinese robot market is driven by high quality, high-value scenarios, and the integration of intelligent technologies, shifting industrial robots from mere programming devices to self-judging equipment [9]. - This evolution allows for the realization of flexible production advantages in multi-category, small-batch manufacturing, leveraging the universal applicability and low entry barriers of AI technology [9].