An LLM-enabled Multi-Agent Autonomous Mechatronics Design Framework

TL;DR

The paper presents an LLM-enabled, multi-agent framework for autonomous mechatronics design that integrates mechanical, electronics, control, and software domains with structured human feedback. A High-Level Planning Agent coordinates domain-specific agents—Mechanical, Simulation & Validation, Electronics, and Embedded Software—to translate system requirements into feasible designs, expressed as $P = f(\bm{F}, \bm{C}, \bm{H})$. The framework is validated on an autonomous water-quality monitoring vessel, demonstrating end-to-end design of a functional propulsion system, validated via CFD/FEA and embedded firmware, with human-guided refinement for complex multiphysics tasks. This approach has the potential to democratize mechatronics design, accelerate cross-disciplinary engineering, and reduce the need for deep domain expertise in physical-system development, while highlighting current limitations in autonomous turbulence modeling and hardware robustness that warrant further research.

Abstract

Existing LLM-enabled multi-agent frameworks are predominantly limited to digital or simulated environments and confined to narrowly focused knowledge domain, constraining their applicability to complex engineering tasks that require the design of physical embodiment, cross-disciplinary integration, and constraint-aware reasoning. This work proposes a multi-agent autonomous mechatronics design framework, integrating expertise across mechanical design, optimization, electronics, and software engineering to autonomously generate functional prototypes with minimal direct human design input. Operating primarily through a language-driven workflow, the framework incorporates structured human feedback to ensure robust performance under real-world constraints. To validate its capabilities, the framework is applied to a real-world challenge involving autonomous water-quality monitoring and sampling, where traditional methods are labor-intensive and ecologically disruptive. Leveraging the proposed system, a fully functional autonomous vessel was developed with optimized propulsion, cost-effective electronics, and advanced control. The design process was carried out by specialized agents, including a high-level planning agent responsible for problem abstraction and dedicated agents for structural, electronics, control, and software development. This approach demonstrates the potential of LLM-based multi-agent systems to automate real-world engineering workflows and reduce reliance on extensive domain expertise.

Figures (5)

• Figure 1: Conceptual architecture of the autonomous mechatronics design framework, illustrating multi-agent collaboration for system development with human-in-the-loop feedback.
• Figure 2: Iterative design process of propellers and hulls generated by the Mechanical Design Agent.
• Figure 3: CFD and structural analysis of the optimized propeller design performed by the Simulation & Validation Agent. (a) Structural stress distribution under laminar flow conditions (baseline analysis). (b) Results obtained from rotating turbulent flow analysis using the Shear Stress Transport (SST) model.Results demonstrate stable hydrodynamic performance, reasonable structural stress levels, and highlight areas for further refinement.
• Figure 4: Electronics Design Agent workflow based on user feedback. The agent analyzes the existing balance car system and reconstructs the core control architecture, including Arduino-based PWM control and H-bridge motor driving
• Figure 5: Embedded control logic and human-verified signal outputs. The agent generates control logic based on the Electronics Design Agent’s architecture, while a logic analyzer is used to capture and verify PWM signals under different scenarios.