The manufacturing industry is an important battlefield for AI large-scale models. The biggest opportunity in the next 10 years lies in the integration of cloud, AI and physical world machines...In the future, intelligent robots that can directly communicate with people will change the shape of the entire factory.
Recently, at the 6th Digital China Construction Summit, Alibaba Chairman and CEO Zhang Yong said so. He revealed that Alibaba Cloud engineers are experimenting with connecting the large model of Tongyi Qianwen to industrial robots, and inputting a sentence of human language in the DingTalk dialog box can remotely command the robot to work.
A water fetching demonstration video released by Alibaba Cloud on the same day showed the results of this experiment. After the engineer sent instructions to the robot through the DingTalk dialog box, the Tongyi Qianwen large model automatically wrote a set of codes and sent them to the robot in the background. The robot began to recognize the surrounding environment, find a bottle of water from a nearby table, and automatically complete the movement A series of actions such as , grabbing, and distribution are delivered to the engineer. This demonstration shows that the Al large model has broken through the limitations of robots, allowing humans to command machines to complete responsible tasks through natural language, allowing robots to truly have a "brain".
According to Alibaba Cloud engineers, AI large models can play an important role in robot development and application stages. For example, during production and operation, the Tongyi Qianwen large model provides the robot with the ability to reason and make decisions. Front-line workers only need to send a paragraph of text, and Tongyi Qianwen can understand its intention and perform task reasoning, and automatically translate it into a machine that can The understood code directs the machine to perform tasks, effectively improving production efficiency.
A1 has shown great potential and value in various fields. In the manufacturing industry, A can improve production efficiency, reduce costs, optimize quality, and enhance innovation capabilities, etc. With the continuous development and progress of A technology, especially the emergence of AI large-scale model technology represented by ChatGPT, the application prospect of AI in the future industry will be broader and deeper.
First of all, let's understand what is an AI large model.
Al large model refers to an artificial intelligence model with ultra-large-scale parameters and data, such as Baidu's Wenxin Yiyan, Alibaba's Tongyi Qianwen, and the global hot ChatGPT. The latest ChatGPT-4 has more than 100 trillion parameter, which is currently the largest natural language processing model in the world. AI large models can process massive amounts of information, generate high-quality content, such as text, images, audio, video, etc., and interact naturally and smoothly with humans. A large model can not only provide more accurate and comprehensive knowledge and services, but also create new knowledge and services, thereby realizing the ability of artificial intelligence generated content (AIGC).
In the author's opinion, the A large model can empower the industry from the perspectives of R&D, design, manufacturing, and management services, and promote the transformation of the manufacturing industry to an intelligent, digital, and networked direction.
R&D design: A large model can be integrated with CAD, EDA, CAE and other industrial design software, and use cloud computing and big data to automatically generate or optimize design solutions to improve design efficiency and accuracy. For example, Cadence has launched Allegro X Al technology, a new generation of system-on-a-chip design technology, which uses generative AI functions to simplify the system design process and shorten the PCB design turnaround time to one-tenth of the original. Manufacturing: By integrating with industrial robots, AGVs, machine tools and other intelligent industrial equipment, AI large models can use technologies such as natural language processing and computer vision to achieve natural interaction and collaboration with humans, improving production efficiency and quality. For example, Siemens combines ChatGPT technology with its existing natural language technology to effectively realize the natural language interaction between the operator and the system; BMW relies on the Nvidia Omniverse platform to optimize the layout of production lines, robots and logistics systems in a virtual way, strengthening the global Factory contact to improve service efficiency.
Management service: A large model can be integrated with industrial management software such as ERP, CRM, MES, etc., and use technologies such as natural language generation and understanding to automatically generate or optimize relevant documents and reports to improve management efficiency and level. For example, Microsoft has launched Dynamics 365 Copilot and Microsoft 365 Copilot, GPT’s interactive AI-capable business products, which have greatly improved users’ work efficiency on business management software. Scenes.
Innovation capability: AI large models can be integrated with industrial Internet platforms and industrial innovation centers, and use technologies such as knowledge graphs and recommendation systems to automatically generate or optimize innovative solutions and services to improve innovation capabilities and value. For example, 4Paradigm has launched an enterprise-level 4Paradigm SageGPT, which integrates large-scale generative language models with vertical domain expertise, has multi-modal and Copoit capabilities in enterprise-level scenarios; Megvii Technology has begun to deploy in the field of large visual models to explore generative AI in warehousing Feasibility in the field of logistics.
It can be said that whether you like it or not, technology A has ushered in a period of great explosion, and technology is developing rapidly. The manufacturing industry should use a positive attitude to view the major changes that the A large model may bring to the manufacturing industry, and make full use of the more opportunities that the AI large model brings to the manufacturing industry, so as to promote the deep integration and coordinated development of the manufacturing industry and other industries , forming a new industrial ecology and competition pattern.
Of course, more AI models will bring more surprises and value to the manufacturing industry, and the author is looking forward to it
