Advancing Machine Learning in Planetary Science

Artificial Intelligence (AI) and Machine Learning (ML) are rapidly transforming planetary science, enabling researchers to analyze increasingly large and complex datasets generated by space missions, ground-based observatories, and remote sensing platforms. As instruments grow more sensitive and mission archives expand into the petabyte scale, traditional manual analysis can no longer keep pace with the volume, dimensionality, and heterogeneity of the data being produced. Established techniques, ranging from unsupervised methods such as clustering and Principal Component Analysis (PCA) for spectral analysis to advanced supervised models including Convolutional Neural Networks (CNNs), are already widely used to study planetary surfaces, atmospheres, interiors, and dynamical systems, while emerging approaches, including Large Language Models (LLMs) and AI agents, are beginning to be explored in planetary-science workflows. These tools not only accelerate discovery but also uncover patterns and correlations that would be difficult or impossible to identify through conventional approaches, opening new avenues for hypothesis generation and testing across the planetary sciences. As the use of AI in planetary science grows, it is also important to understand the ethical responsibilities of employing machine learning. In this rapidly evolving AI landscape, clarity and transparency regarding data usage and methods are critical.This Research Topic aims to bring together contributions that showcase the breadth and depth of ML applications across planetary research. We welcome original research articles, reviews, and methodological papers addressing, but not limited to, the following areas:• Surface and subsurface characterization: detection, segmentation, and geospatial processing of compositional and morphological features across planetary surfaces and subsurface environments, including automated mapping of geological units, mineralogical classification, and feature extraction from imagery and spectral data.• Atmospheric and climate studies: ML-driven analysis of atmospheric composition, circulation patterns, cloud dynamics, and climate modelling across planetary bodies, including Earth as a comparative case study, as well as retrievals, radiative-transfer calculations, and parameter estimation for planetary and exoplanetary atmospheres.• Planetary system formation, exoplanets, and astrobiology: applications of ML to the detection and characterization of exoplanets, protoplanetary disk studies, biosignature identification, and models of planetary system evolution.• Small body dynamics and physical properties: ML approaches to characterizing the composition, shape, rotation state, and orbital dynamics of asteroids, comets, and other minor bodies in the Solar System.• Autonomous and onboard AI for planetary exploration: ML methods for navigation, target selection, onboard data processing and prioritization, and autonomous rover and spacecraft operations.• Data integration, fusion, and dissemination: methods for combining multi-instrument, multi-mission, and multi-modal datasets, along with tools for visualization of results.• LLMs and AI agents in scientific workflows: emerging use cases for language models and autonomous agents in literature synthesis, hypothesis generation, code development, mission planning, and other stages of the planetary science research pipeline.• Ethics and responsible use of AI: transparency, reproducibility, and accountability in ML-based planetary science, including data provenance, bias and reliability of AI-derived results, and best practices for the responsible development and deployment of AI methods.By highlighting both established and emerging ML methodologies, this Research Topic seeks to foster interdisciplinary dialogue between planetary scientists and data scientists, and to support the responsible and effective integration of AI into the future of planetary exploration. Topic editors: Stavro Ivanovski, Yann Alibert Source: https://www.frontiersin.org/research-topics/85369/advancing-machine-learning-in-planetary-science