A year ago, the automation specialists from Beckhoff in Verl explored the possibilities of Large Language Models (LLM), which are still in their infancy, and presented an application in the automation sector with TwinCAT Chat. At HANNOVER MESSE 2024, the East Westphalian company will be demonstrating what has been achieved in the twelve months since then.

Advantages from the programmer's point of view

But first, a brief digression on the benefits that developers expect from LLMs, previously known primarily through ChatGPT. According to Beckhoff, the language models offer a number of advantages for both automation engineers and company management. For automation engineers, LLMs have the potential to revolutionize the development process by automatically generating and completing code - and thus speeding up the entire process. In addition, LLMs can even create personal tutorials and ask for specific solutions to problems that arise, which also speeds up the process of finding solutions. Another advantage is the ability to consistently implement and adhere to guidelines and best practices in the field of automation.

Advantages from the perspective of company management

From a company management perspective, LLMs promote the transfer of knowledge within the company. They can act as a central knowledge database that stores valuable information and makes it available when required. LLMs can also relieve the burden on support by acting as the first point of contact for customer inquiries. This not only improves response times, but should also lead to greater customer satisfaction. Overall, LLMs offer an efficient and innovative solution for numerous challenges in the modern business world.

LLMs as a tool for programmers, not their replacement

According to Beckhoff, the aim is not to completely replace programmers, nor does current technical developments suggest that this is the case. Rather, the aim is to provide programmers with ever better tools so that they can work effectively. The aim is to increase the productivity of programmers - not least as one of the main ways of combating the shortage of skilled workers. If vacancies cannot be filled because there are simply no qualified specialists to be found, AI can be used to ensure continued competitiveness.

Equipped with the expertise of the automation industry

The development of TwinCAT Chat was based on the idea of offering users a clear advantage over the use of conventional LLMs, for example ChatGPT in the web browser. The decisive added value lies in its deep integration, especially with regard to the specialized requirements of the automation industry. The core functions of TwinCAT Chat include integration directly into the development environment (IDE). This simplifies the development process considerably, as communication and code exchange are seamlessly integrated. In addition, the model has been initialized specifically for TwinCAT queries. This means you can ask your specific questions directly and do not have to tell the model that you are using TwinCAT and expect the code examples in structured text.

Efficient, fast and fewer errors

Another highlight is the ability to simply transfer generated code. This not only saves developers time, but also reduces human errors that can occur during manual transfer. Interaction with TwinCAT Chat has been designed in such a way that typing commands is reduced to a minimum. Instead, requests pre-tested by Beckhoff that are specifically designed to improve the user's workflow can be used simply by clicking the mouse. In addition to PLC code generation, Beckhoff is also working on a chatbot that automatically creates a TwinCAT HMI project. The aim is that a user only has to formulate how they would like their HMI to be structured and TwinCAT generates the complete HMI project in the background. The customer therefore receives immediate feedback in the form of the visualized HMI. Another project involves a chatbot interface to the Beckhoff documentation system. Beckhoff will inform interested visitors about the exact status of current developments at HANNOVER MESSE 2024.

Video