EUPHORIA: Efficient Universal Planning via Hybrid Optimization for Robust Industrial Robotic Assembly

2026-05-15

AI Summary

Existing robotic assembly methods for construction either lack generalization due to high task-specificity or suffer from inefficiency by decoupling structural sequencing from motion planning. This work proposes EUPHORIA, a unified framework that integrates structural reasoning and motion planning to enable few-shot generalization and dynamically efficient execution. Key innovations include a graph hypernetwork-based meta-geometric encoder for rapid, zero-gradient, parameter-level adaptation; a physics-informed Graph Transformer with contact-force-modulated attention; and differentiable residual stability correction to bridge the Sim2Real gap. Experiments demonstrate that the method achieves state-of-the-art success rates on unseen, non-standard geometric structures with minimal training samples, while significantly reducing energy consumption and consistently outperforming decoupled baselines.

Abstract

Robotic assembly in architectural construction faces a persistent bottleneck: existing planners are either highly specialized, requiring prohibitive retraining for every new geometric design, or operationally inefficient, treating structural sequencing and kinematic motion as disjoint processes. We present EUPHORIA, a unified framework that achieves universal few-shot adaptability and dynamic efficiency through a hybrid optimization strategy. To overcome the retraining bottleneck, we propose a Meta-Geometric Encoder based on Graph Hypernetworks: unlike standard contrastive learning, which performs only feature-level recognition, our hypernetwork dynamically generates policy parameters from a minimal support set, enabling parameter-level adaptation to complex topologies (e.g., domes, arches) without gradient-based retraining. For structural reasoning, we introduce a Physics-Informed Graph Transformer trained via Soft Actor-Critic (SAC), with a Physics-Bias Attention mechanism that modulates attention scores using contact forces from Discrete Element Model (DEM) simulations, guiding the planner toward structurally critical connections. We further ensure operational efficiency through Kinematics-Aware Sequencing, where the SAC objective penalizes high-energy transitions. Finally, we bridge the Sim2Real gap via Residual Stability Correction, a differentiable optimization layer that fine-tunes coarse assembly actions by minimizing a joint energy-stability cost prior to execution. Experiments show that EUPHORIA significantly reduces energy consumption over decoupled baselines and achieves state-of-the-art success rates on unseen, non-standard geometries with minimal few-shot examples, fusing meta-learning, physics-informed attention, and residual optimization into a cohesive, generalized planner.

Problem

• robotic assembly
• universal planning
• geometric generalization
• structural sequencing
• kinematic efficiency

Innovation

• Meta-Geometric Encoder
• Physics-Informed Graph Transformer
• Kinematics-Aware Sequencing
• Residual Stability Correction
• Hybrid Optimization