Astronomy16 papers
Artificial Intelligence and Machine Learning in Planetary Science
A companion collection for the ACM 2026 talk by Valerio Carruba
VC
Valerio Carruba
All Reviews
AstronomyMust Read
intermediate
Deep operator neural network applied to efficient computation of asteroid surface temperature and the Yarkovsky effect
Zhao, Shunjing et al. (2024)
- Published
- Nov 1, 2024
- Journal
- Astronomy & Astrophysics · Vol. 691
- DOI
- 10.1051/0004-6361/202451789
At a GlanceAI
DeepONet predicts asteroid surface temperatures ~100,000 times faster while preserving ~1% accuracy for Yarkovsky studies.
SummaryAI
Computing asteroid surface temperatures accurately is essential for understanding thermal properties and orbital changes, but repeated high-fidelity simulations are often too slow. This study applies a deep operator neural network to emulate those temperature calculations, achieving about 1% average error at a computational cost five orders of magnitude lower. The speedup makes broad multidimensional thermal-parameter studies much more practical. As an initial demonstration, the model is used to estimate Yarkovsky-driven orbital evolution for asteroids including Phaethon and 2001 WM41.
Method SnapshotAI
The authors use a deep operator neural network surrogate model for asteroid thermophysical calculations and embed it in N-body orbital simulations.
BackgroundAI
General familiarity with asteroid thermophysics, machine learning surrogate models, and orbital dynamics is helpful.
A pioneering application of DeepONets to asteroid thermophysical modeling, this work shows how neural operator architectures can dramatically accelerate surface temperature calculations while preserving high accuracy. Its methodology paves the way for efficient coupling of AI-based thermal models with orbital dynamics, making it an important contribution to the growing use of scientific machine learning in planetary science.
— VC