AB-UPT for Automotive and Aerospace Applications

2025-10-17

AI Summary

High computational cost hinders practical deployment of fluid dynamics simulations in automotive and aerospace engineering. To address this, we propose the Anchored-Branch Universal Physics Transformer (AB-UPT), a lightweight neural surrogate model designed for multi-geometry aerodynamic performance prediction. AB-UPT explicitly encodes physical laws via anchored geometric priors and branch-wise physics constraints, enabling efficient training on a single GPU. Evaluated on high-fidelity SHIFT-SUV and SHIFT-Wing datasets generated via Luminary Cloud, AB-UPT achieves second-level, high-accuracy aerodynamic force prediction (error < 0.5%) in real-world scenarios, with inference latency reduced by one to two orders of magnitude over existing Transformer-based baselines. This work pioneers the integration of the anchored-branch architecture into universal physics-informed modeling, advancing the development of industrial-grade, deployable neural solvers.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Machine Learning: Deep Neural Architectures and Foundation Models
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

The recently proposed Anchored-Branched Universal Physics Transformers (AB-UPT) shows strong capabilities to replicate automotive computational fluid dynamics simulations requiring orders of magnitudes less compute than traditional numerical solvers. In this technical report, we add two new datasets to the body of empirically evaluated use-cases of AB-UPT, combining high-quality data generation with state-of-the-art neural surrogates. Both datasets were generated with the Luminary Cloud platform containing automotives (SHIFT-SUV) and aircrafts (SHIFT-Wing). We start by detailing the data generation. Next, we show favorable performances of AB-UPT against previous state-of-the-art transformer-based baselines on both datasets, followed by extensive qualitative and quantitative evaluations of our best AB-UPT model. AB-UPT shows strong performances across the board. Notably, it obtains near perfect prediction of integrated aerodynamic forces within seconds from a simple isotopically tesselate geometry representation and is trainable within a day on a single GPU, paving the way for industry-scale applications.

Problem

• Replicate automotive CFD simulations efficiently
• Evaluate AB-UPT on automotive and aerospace datasets
• Achieve accurate aerodynamic predictions with minimal computation

Innovation

• AB-UPT uses neural surrogates for fluid dynamics simulations
• It predicts aerodynamic forces from tessellated geometry representation
• Model trains in one day on a single GPU