Recent Advances in PLC Programming Using Artificial Intelligence and Large Language Models

Abstract

Integrating Artificial Intelligence (AI) into process control is one of the most significant technological trends in industrial automation today. The generative programming of Programmable Logic Controllers (PLCs) is a prominent example of this development. While Artificial Neural Networks (ANNs) have previously been applied for tasks such as natural language processing and fault prediction in PLC hardware, recent advancements in Large Language Models (LLMs) have further expanded AI capabilities, enabling the interpretation of complex prompts and assisting with control code generation. Development tools and industrial copilots powered by generative AI are increasingly being proposed to support engineers in managing control systems, with the potential to simplify and accelerate control software development. In contrast to these promises, using generative models in PLC programming is still in its early stages, characterized by exploratory research and cautious implementation. This review provides a systematic overview of recent developments in AI-assisted PLC programming, focusing on generative approaches. It synthesizes emerging methodologies, tools, and applications while critically examining current limitations and outlining potential research directions in industrial control systems.