From Code Generator to Engineering Companion: A Plant-aware Agent for PLC Programming

Joschka Kersting; Gesa Benndorf

cs.SEOct 20, 2026

From Code Generator to Engineering Companion: A Plant-aware Agent for PLC Programming

Authors: Joschka Kersting, Gesa Benndorf

Organizations: Fraunhofer IOSB-INA Lemgo, Germany

Abstract

Programmable Logic Controller (PLC) coding remains a manual, IDE-centric task that demands constant context switching between documentation, project files, and code editors. Recent work has demonstrated that retrieval-augmented large language models can generate vendor-compliant Structured Text at high compilation rates \cite{pred}, yet these systems treat code generation as a one-shot pipeline: the engineer submits a query and receives code, with no persistent knowledge of the specific plant. We present a plant-aware conversational copilot that extends this paradigm in three directions: plant-specific context ingestion, a chat-first interaction model for explanation and debugging, and a graduated tool-access ladder from chat to deployment. The system is designed for real plant infrastructure and keeps the engineer in control of each step that produces real-world effects.

Figures & tables

System	Plant context	Code gen	Compile	Deploy	Chat-first	Multi-vendor	Target platform(s)
LLM4PLC [ 2 ]	–	✓	✓	–	–	–	Siemens SCL
Agents4PLC [ 3 ]	–	✓	✓	–	–	✓	IEC 61131-3 (generic)
AutoPLC [ 4 ]	–	✓	✓	–	–	✓	Siemens SCL, CODESYS ST
Spec2Control [ 5 ]	–	✓	–	✓	–	–	ABB
Siemens Copilot [ 6 ]	–	✓	✓	✓	✓	–	Siemens TIA
Prior work [ 1 ]	–	✓	✓	–	–	–	Mitsubishi iQ-R

TABLE I: Comparison of related systems

Fig. 1: The prototype web application used to control generation, compilation, and plant-aware context. The left panel exposes generation settings and MCP actions, while the main pane shows the conversation and generated ST code.

Panel of Figure 3 — Fig. 2: System architecture for our ST coding assistant with plant-aware retrieval, user-triggered compilation and PLC control via MCP (based on [ 1 ] ).

Fig. 3: Demonstration plant setup at a research test facility providing metal discs in different colors, inserts, sensors, and actuators for sorting them. The plant is equipped with a Mitsubishi Electric iQ-R series CPU (model R04).

Coding Agents

Coding agents use language models, tools, and feedback to modify software and complete engineering tasks.

See all 139 papers

PLCWorld: Benchmarking LLM-Generated PLC Programs in Closed-Loop Plant Simulation
Oct 2, 2026Yunji Kim, Yunseok Lee, Hyunwoo Seo +2LLM-Based Program Synthesis Code Generation
Dongguk University · MOAI technologies
Exploring the Agentic Frontier of Verilog Code Generation
Mar 19, 2026Patrick Yubeaton, Siddharth Garg, Chinmay HegdeCode Generation Agentic Code Generation
New York University, Tandon School of Engineering, Brooklyn, NY
Retrieval-grounded robot program generation and simulation-based correction via Model Context Protocol
Aug 13, 2026Zhichao Zhou, Siyuan Chen, Omkar Salunkhe +3Code Generation for Robotics LLM-Based Program Synthesis
Department of Mechanical Engineering, Chalmers University of Technology, Hörsalsvägen 7a, Gothenburg, SE-412 96, Sweden

Oct 2, 2026J/K move · Enter open · S save

PLCWorld: Benchmarking LLM-Generated PLC Programs in Closed-Loop Plant Simulation

Yunji Kim, Yunseok Lee, Hyunwoo Seo, Jaerim Choi+1

Dongguk University · MOAI technologies

LLM-Based Program Synthesis Code Generation LLM Safety Benchmarks Feedback Control

Open paper AskPDF arXiv

Programmable logic controllers (PLCs) coordinate industrial equipment by reading sensor inputs and issuing control commands. Evaluating whether large language model (LLM)-generated PLC programs satisfy task requirements and safety constraints requires observing how their commands affect device and workpiece states. We introduce PLCWorld, a common closed-loop execution environment and benchmark that couples Structured Text (ST) execution with simulated plant responses and sensor feedback. Grounded in control relations identified in industrial PLC programs and engineering documentation, PLCWorld contains 100 synthetic tasks and 473 registered task-condition pairs across Motion Control and Material Handling, with difficulty defined by control-dependency scope. A common protocol reports Task Success and Safety Violation separately. Validation combines practitioner review, reference and alternative programs, targeted counterexamples, specification-evaluator alignment checks, and comparisons with independent ST runtimes. Reference and alternative programs satisfy their applicable cases, while all 542 targeted counterexamples activate their designated evaluator rules under at least one registered condition. Execution Gap relates submission-profile acceptance to subsequent task failure or observed Safety Violation. Across the constructed task groups, direct GPT-5.5 achieves 82.70% Task Success on Easy cases but 25.10% on Hard cases. Evaluations of six LLMs and four adapted generation-and-verification workflows further expose differences between completion, safety, and generation cost. Our code, simulation environment, benchmark tasks, and baseline implementations are publicly available at https://yunji0516.github.io/PLCWorld/.

Abstract

Coding Agents

Explore similar work

PLCWorld: Benchmarking LLM-Generated PLC Programs in Closed-Loop Plant Simulation

Exploring the Agentic Frontier of Verilog Code Generation

Retrieval-grounded robot program generation and simulation-based correction via Model Context Protocol