AgenticControl: An Automated Control Design Framework Using Large Language Models

TL;DR

AgenticControl leverages multi-agent large language models to automate control system design, improving robustness and performance across diverse systems through structured collaboration and adaptive optimization.

Contribution

This work introduces a multi-agent LLM framework for automated control design, integrating structured communication, robust optimization, and real-time adaptability, validated on multiple control systems.

Findings

Full State Feedback matches LQR performance

PID controller outperforms MATLAB's PIDTuner by 55%

DeepSeek achieves fastest convergence among tested LLMs

Abstract

Traditional control system design, reliant on expert knowledge and precise models, struggles with complex, nonlinear, or uncertain dynamics. This paper introduces AgenticControl, a novel multi-agent framework that automates controller design using coordinated Large Language Model (LLM) agents. Through structured JSON communication, these agents handle tasks including controller selection, scenario design, parameter optimization, performance evaluation, and decision-making. Through an actor-critic optimization approach, the system iteratively improves performance while progressing through scenarios of increasing complexity to ensure robustness under nominal conditions, measurement noise, actuator disturbances, and parametric uncertainties. Key innovations include structured multi-agent collaboration, robust optimization mechanisms, and real-time adaptability via in-context learning. Validated across four diverse control systems, namely, DC Motor Position control, Ball and Beam, Inverted Pendulum, and Double Inverted Pendulum, the framework achieves competitive performance against classical methods. Its Full State Feedback solution closely matches Linear Quadratic Regulator (LQR) results, while the designed PID controller significantly outperforming MATLAB's PIDTuner, reducing PID tracking error by 55% through adaptive parameter exploration. A comparative study of five LLM models reveals distinct optimization profiles, with DeepSeek achieving the fastest convergence. This work demonstrates the potential of LLM-driven control design, paving the way for advanced techniques like model predictive control and reinforcement learning.