# From Detection to Action: Using LLM Agents for Fault-Tolerant Control

arXiv: https://arxiv.org/abs/2606.28011 · HTML: https://mlsanity.com/paper/2606.28011/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use.
- Builds on 15 of its 32 references that are in the map.
- First posted on arXiv 26 Jun 2026.

## Placed in

- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)

## Builds on (most cited first, 15 of 15)

- [SOSA: A Lightweight Ontology for Sensors, Observations, Samples, and Actuators](https://mlsanity.com/paper/1805.09979/) (2018)
- [Autonomous Industrial Control using an Agentic Framework with Large Language Models](https://mlsanity.com/paper/2411.05904/) (2024)
- [Pre-Trained Large Language Models for Industrial Control](https://mlsanity.com/paper/2308.03028/) (2023)
- [An Agentic Approach to Automatic Creation of P&ID Diagrams from Natural Language Descriptions](https://mlsanity.com/paper/2412.12898/) (2024)
- [Control Industrial Automation System with Large Language Model Agents](https://mlsanity.com/paper/2409.18009/) (2024)
- [Control-Informed Reinforcement Learning for Chemical Processes](https://mlsanity.com/paper/2408.13566/) (2024)
- [Leveraging LLM Agents and Digital Twins for Fault Handling in Process Plants](https://mlsanity.com/paper/2505.02076/) (2025)
- [AutoChemSchematic AI: Agentic Physics-Aware Automation for Chemical Manufacturing Scale-Up](https://mlsanity.com/paper/2505.24584/) (2025)
- [Systematic Comparison of Software Agents and Digital Twins: Differences, Similarities, and Synergies in Industrial Production](https://mlsanity.com/paper/2307.08421/) (2023)
- [Integrating Ontology Design with the CRISP-DM in the context of Cyber-Physical Systems Maintenance](https://mlsanity.com/paper/2407.06930/) (2024)
- [Knowledge Graph Modeling-Driven Large Language Model Operating System (LLM OS) for Task Automation in Process Engineering Problem-Solving](https://mlsanity.com/paper/2408.14494/) (2024)
- [Representing Time-Continuous Behavior of Cyber-Physical Systems in Knowledge Graphs](https://mlsanity.com/paper/2506.13773/) (2025)
- [Benchmark for Planning and Control with Large Language Model Agents: Blocksworld with Model Context Protocol](https://mlsanity.com/paper/2512.03955/) (2025)
- [A Reinforcement Learning-based Economic Model Predictive Control Framework for Autonomous Operation of Chemical Reactors](https://mlsanity.com/paper/2105.02656/) (2021)
- [Chatbot-Based Ontology Interaction Using Large Language Models and Domain-Specific Standards](https://mlsanity.com/paper/2408.00800/) (2024)

## Built on by (most cited first, 2 of 2)

- [ADMITBench: A Safety-Governed Reference Framework for Evaluating the Admissibility of Industrial LLM Advisories](https://mlsanity.com/paper/2608.03866/) (2026)
- [LLM-Guided Contextual Action Evaluation for Operational Decisions in Industrial Processes](https://mlsanity.com/paper/2608.24156/) (2026)

## Abstract (arXiv)

We propose an agentic Large Language Model (LLM) framework for active Fault-Tolerant Control (FTC) that transforms fault detection outputs into constraint-aware recovery actions grounded in plant-specific knowledge. The approach couples (i) a multi-agent workflow that decomposes operator duties into monitoring, planning, action synthesis, simulation, validation, and reprompting; (ii) a Digital Process Plant Twin (DPPT) that exposes plant data, models, and a simulation service for pre-execution testing; and (iii) a Graph Retrieval-Augmented Generation (Graph RAG) layer built on the CPSMod ontology, which organizes plant knowledge (structure, function, hybrid dynamics, control context, and fault semantics) into a graph that supports relation-aware, multi-hop retrieval for the agents. Corrective actions are generated as minimal-risk state-machine recovery paths and corresponding discrete commands or continuous setpoint adaptations, then validated deterministically against interlocks, envelopes, and dynamic feasibility before any actuation. If no acceptable plan is found within a bounded time window, control is handed to a safety fallback. The framework is evaluated in simulation on two representative benchmarks: a discrete batch Mixing Module and a Continuous Stirred-Tank Reactor (CSTR) under closed-loop PID regulation. Results with lightweight LLMs (GPT-4o-mini and GPT-4.1-mini) show that semantically grounded agents can derive valid recovery decisions within latency budgets compatible with the respective process dynamics, demonstrating a practical pathway from detection to validated corrective action across both discrete and continuous FTC tasks.