PSTNet: Physically-Structured Turbulence Network

2026-03-09

AI Summary

This study addresses the challenge of accurately estimating atmospheric turbulence intensity in sparsely observed regions such as oceans and polar areas, where real-time airborne estimation remains difficult. To this end, the authors propose a lightweight neural network architecture comprising only 552 learnable parameters. The method uniquely integrates physical priors—including Monin–Obukhov similarity theory, Richardson-number-derived soft labels, density-ratio modulation, and Kolmogorov’s inertial subrange scaling law—into the model design. This integration is achieved through a zero-parameter physics-based backbone, mixture-of-experts subnetworks supervised by soft targets, Feature-wise Linear Modulation layers, and an output layer enforcing the Kolmogorov scaling law, thereby enabling deep fusion of physical consistency and data-driven learning. Evaluated on 340 hypersonic flight simulations, the approach reduces average miss distance by 2.8%, achieves a win rate of 78%, and executes inference within 12 seconds on a Cortex-M7 microcontroller, meeting the stringent requirements of airborne safety-critical systems.

Technology Category

• Intelligent Robots: State Estimation
• Search and Optimization: Learning to Search
• Machine Learning: Mixture of Experts (MoE)

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Reliable real-time estimation of atmospheric turbulence intensity remains an open challenge for aircraft operating across diverse altitude bands, particularly over oceanic, polar, and data-sparse regions that lack operational nowcasting infrastructure. Classical spectral models encode climatological averages rather than the instantaneous atmospheric state, and generic ML regressors offer adaptivity but provide no guarantee that predictions respect fundamental scaling laws. This paper introduces the Physically-Structured Turbulence Network (PSTNet), a lightweight architecture that embeds physics directly into its structure. PSTNet couples four components: (i) a zero-parameter backbone derived from Monin-Obukhov theory, (ii) a regime-gated mixture of specialist sub-networks supervised by Richardson-number-derived soft targets, (iii) Feature-wise Linear Modulation layers conditioning hidden representations on local air-density ratio, and (iv) a Kolmogorov output layer enforcing inertial-subrange scaling as an architectural constraint. The entire model contains only 552 learnable parameters, requiring fewer than 2.5 kB of storage and executing in under 12s on a Cortex-M7 microcontroller. We validate PSTNet on 340 paired six-degree-of-freedom guidance simulations spanning three vehicle classes (Mach 2.8, 4.5, and 8.0) and six operational categories with real-time satellite weather ingestion. PSTNet achieves a mean miss-distance improvement of +2.8% with a 78% win rate and a statistically significant effect size. Our results demonstrate that encoding domain physics as architectural priors yields a more efficient and interpretable path to turbulence estimation accuracy than scaling model capacity, establishing PSTNet as a viable drop-in replacement for legacy look-up tables in resource-constrained, safety-critical on-board guidance systems.

Problem

• atmospheric turbulence
• real-time estimation
• data-sparse regions
• nowcasting
• turbulence intensity

Innovation

• physically-informed neural network
• turbulence estimation
• architectural priors
• Kolmogorov scaling
• resource-constrained inference