Towards LLM-Assisted Architecture Recovery for Real-World ROS~2 Systems: An Agent-Based Multi-Level Approach to Hierarchical Structural Architecture Reconstruction

2026-05-19

AI Summary

This work addresses the challenge of recovering the hierarchical software architecture of ROS 2 systems, which is implicitly embedded in distributed artifacts and difficult to reconstruct explicitly. Existing approaches lack support for decomposing and composing structures across multiple abstraction levels. To overcome this limitation, the paper proposes an agent-based, multi-level architecture recovery method that integrates blueprint-guided reasoning with large language models (LLMs). By analyzing launch file dependencies and parsing node lists, the approach leverages multi-granular intermediate architectural representations and refined prompting strategies to iteratively reconstruct cross-layer structural constraints. Experimental evaluation on a real-world collaborative robotic disassembly system demonstrates that the method significantly improves structural consistency, scalability, and robustness in architecture recovery, while also uncovering critical challenges related to dynamic semantic integration in large-scale systems.

Abstract

Explicit software architecture models are essential artifacts for communicating, analyzing, and evolving complex software-intensive systems. In ROS~2-based robotic systems, however, structural (de-)composition and integration semantics are often only implicitly encoded across distributed artifacts such as source code and launch files, making recovery of hierarchical architecture particularly difficult. Existing approaches mainly focus on node-level entities and communication wiring, while providing limited support for recovering hierarchical structural (de-)composition across multiple abstraction levels. In this paper, we extend our previously proposed blueprint-guided LLM-assisted architecture recovery pipeline for ROS~2 systems through two major enhancements: (1) refined prompting to improve the consistency and controllability of architecture synthesis, and (2) a staged recovery strategy based on multi-level intermediate architectural representations that incorporate the atomic ROS node list and launch file dependencies, thereby enabling structurally constrained reconstruction across multiple abstraction levels. The approach is evaluated on a real-world automated product disassembly system based on cooperative robotic arms and heterogeneous ROS~2 artifacts. Compared to our previous work, the considered case study exhibits substantially higher integration complexity and richer functionality. The results demonstrate improved structural consistency, scalability, and robustness of architecture recovery, while also revealing remaining challenges related to dynamic integration semantics in large-scale ROS~2 systems.

Problem

• architecture recovery
• ROS 2
• hierarchical structure
• software architecture
• structural decomposition

Innovation

• LLM-assisted architecture recovery
• multi-level architecture reconstruction
• ROS 2
• agent-based approach
• hierarchical structural decomposition