From AI-assisted to AI-native Industrial Control: Architecture, Agent Design, and Challenges
Abstract
The integration of AI and industrial control is rapidly transforming industrial automation. However, current approaches remain predominantly AI-assisted, where AI functions are incorporated as auxiliary components without fundamentally altering the underlying architecture of industrial control systems. Consequently, the potential of AI to enable higher levels of autonomy, adaptability, and continuous evolution remains underutilized. This paper proposes an AI-native industrial control paradigm, which reconceptualizes industrial control systems by positioning AI as the intrinsic core rather than an external enhancement. The proposed paradigm is built upon an AI-native infrastructure and a collection of collaborative agentic AIs, each encapsulating a set of advanced skills. Among them, the industrial control agent plays a central role in orchestrating the system lifecycle. Through this paradigm, industrial control shifts from a closed, capability-oriented, human-driven, and passive engineering process to an open, task-oriented, intent-driven, and autonomous solution delivery mechanism. The paper further analyzes the key challenges of the AI-native paradigm from both AI and industrial control perspectives, including issues related to AI model reliability, formal verification, real-time assurance, and heterogeneous resource orchestration. Overall, this work outlines a forward-looking vision for industrial control in the AI era and identifies critical research directions toward autonomous, evolvable, and adaptive control systems in complex and dynamic industrial environments.
Related
- A quantitative model for embedding quantum-secure communications and evaluating communication autonomy in industrial control systems
- On some challenges of open-loop wake steering control: a row of 4 offshore wind turbines investigated with LES
- Plasmon-assisted coherent control of a four-level semiconductor quantum dot: operating window and fidelity limits
- Spectral-control-assisted Kerr-lens mode-locking in a Tm:Y2O3 ceramic laser
- Accurate gap control of 3D-integrated superconducting quantum circuits using asymmetric indium architecture and multi-angle flip-chip bonding
BibTeX
@misc{liu2026from,
title = {{From AI-assisted to AI-native Industrial Control: Architecture, Agent Design, and Challenges}},
author = {Lisi Liu and Bin Wang and Quan Tang and Chuan Pang and Feng Gao and Shengkai Liu},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.34773},
eprint = {2610.34773},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.34773},
}
RIS
TY - UNPB
TI - From AI-assisted to AI-native Industrial Control: Architecture, Agent Design, and Challenges
AU - Lisi Liu
AU - Bin Wang
AU - Quan Tang
AU - Chuan Pang
AU - Feng Gao
AU - Shengkai Liu
PY - 2026
DA - 2026/10/26
PB - arXiv
DO - 10.48550/arxiv.2610.34773
UR - https://arxiv.org/abs/2610.34773
AB - The integration of AI and industrial control is rapidly transforming industrial automation. However, current approaches remain predominantly AI-assisted, where AI functions are incorporated as auxiliary components without fundamentally altering the underlying architecture of industrial control systems. Consequently, the potential of AI to enable higher levels of autonomy, adaptability, and continuous evolution remains underutilized. This paper proposes an AI-native industrial control paradigm, which reconceptualizes industrial control systems by positioning AI as the intrinsic core rather than an external enhancement. The proposed paradigm is built upon an AI-native infrastructure and a collection of collaborative agentic AIs, each encapsulating a set of advanced skills. Among them, the industrial control agent plays a central role in orchestrating the system lifecycle. Through this paradigm, industrial control shifts from a closed, capability-oriented, human-driven, and passive engineering process to an open, task-oriented, intent-driven, and autonomous solution delivery mechanism. The paper further analyzes the key challenges of the AI-native paradigm from both AI and industrial control perspectives, including issues related to AI model reliability, formal verification, real-time assurance, and heterogeneous resource orchestration. Overall, this work outlines a forward-looking vision for industrial control in the AI era and identifies critical research directions toward autonomous, evolvable, and adaptive control systems in complex and dynamic industrial environments.
ER -