A Neuro-Symbolic Framework for Vision-Language-Action Deployment in Industrial Assembly

This paper proposes a novel neuro-symbolic framework that augments Vision-Language-Action (VLA) models to assist high-precision operations in industrial assembly. While Large Language Models (LLMs) demonstrate expertise in interpreting natural language instructions, they often struggle with processing structured manufacturing data and lack long-horizon planning capabilities and deterministic physical awareness required for complex robotic execution. Furthermore, their susceptibility to generating physically unfeasible sequences and their lack of robust error-recovery mechanisms limit their autonomous deployment. To address these limitations, the proposed pipeline decomposes the assembly process into sequential stages of planning, technological grounding, adaptive execution, and recovery. Initially, the framework leverages LLMs to synthesize multimodal inputs—ranging from natural language directives to formalized product and process specifications—into a formal Hierarchical Task Network (HTN). To ensure physical feasibility, an external symbolic planner rigorously validates the HTN’s causal dependencies against established spatial and operational constraints. This verified logical sequence is subsequently compiled into a reactive Behavior Tree (BT) to guide the VLA model. During execution, a real-time object detection system and an atomic skill library dynamically translate the BT into grounded actions. Designed for human-robot collaborative environments, the architecture autonomously allocates standard operations to the robotic agent while safely deferring tasks requiring high dexterity to human operators. By combining the semantic adaptability of LLMs with the rigorous state validation of symbolic planners, this methodology ensures deterministic, safe, and efficient manufacturing workflows. The proposed architecture is evaluated against both physical and logical criteria using task-specific evaluation metrics.