Real-time design of architectural structures with differentiable mechanics and neural networks

2024-09-04

AI Summary

Traditional geometric design of architectural structures (e.g., masonry shells, cable-net towers) relies on computationally expensive iterative optimization, hindering rapid exploration and real-time design iteration. Method: This paper proposes a real-time design framework integrating differentiable mechanical simulation with neural networks. Its core innovation is the “differentiable mechanical constraint embedding” paradigm: physics-informed, differentiable rigid-body and truss-based mechanical models are embedded as structural priors into an end-to-end neural network, enabling joint optimization of geometry generation and mechanical integrity. Contribution/Results: The method achieves target geometry with accuracy comparable to conventional optimization while guaranteeing mechanical feasibility—yielding real-time inference speeds and superior generalization over purely data-driven approaches. It has been integrated into mainstream 3D modeling software and validated via physical prototypes.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Deep Neural Architectures and Foundation Models
• Humans and AI: Game Design — Procedural Content Generation & Storytelling

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

Designing mechanically efficient geometry for architectural structures like shells, towers, and bridges is an expensive iterative process. Existing techniques for solving such inverse mechanical problems rely on traditional direct optimization methods, which are slow and computationally expensive, limiting iteration speed and design exploration. Neural networks would seem to offer a solution, via data-driven amortized optimization for specific design tasks, but they often require extensive fine-tuning and cannot ensure that important design criteria, such as mechanical integrity, are met. In this work, we combine neural networks with a differentiable mechanics simulator to develop a model that accelerates the solution of shape approximation problems for architectural structures modeled as bar systems. As a result, our model offers explicit guarantees to satisfy mechanical constraints while generating designs that match target geometries. We validate our model in two tasks, the design of masonry shells and cable-net towers. Our model achieves better accuracy and generalization than fully neural alternatives, and comparable accuracy to direct optimization but in real time, enabling fast and sound design exploration. We further demonstrate the real-world potential of our trained model by deploying it in 3D modeling software and by fabricating a physical prototype. Our work opens up new opportunities for accelerated physical design enhanced by neural networks for the built environment.

Problem

• Accelerates design of architectural structures using neural networks.
• Ensures mechanical integrity while matching target geometries.
• Enables real-time design exploration for shells, towers, and bridges.

Innovation

• Combines neural networks with differentiable mechanics
• Ensures mechanical integrity in real-time design
• Integrates into 3D modeling for rapid prototyping