← Back to all news
arXiv Machine Learning By Ana Henrique, Sofia Sheikh, Benjamin Fields

Solar System Surface Technosignatures: Searching for Signs of Artificial Design on Extraterrestrial Terrain

Read the original on arXiv Machine Learning →

The paper proposes expanding technosignature searches beyond radio emissions to include surface technosignatures—intentional, non-random patterns on planetary terrain. It argues that on geologically inactive bodies like the Moon and Mars, such artificial traces could persist for hundreds of millions of years and can be detected using high-resolution imaging (e.g., HiRISE) combined with machine‑learning anomaly‑detection. The authors outline supervised and unsupervised modeling approaches, and suggest that complementary LiDAR, radar, and spectroscopy could provide quantitative limits on surface technosignatures while benefiting planetary science.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

Hugging Face Trending Papers
Aug 10

A Machine Learning Based Search for Lunar Anomalies

The Lunar Reconnaissance Orbiter (LRO) has been collecting high-resolution images (at around 0. 5-2 meters per pixel linearly with its Narrow Angle Camera) of the Moon since 2009, amassing a large dataset of images and offering researchers the opportunity to study the surface of the Moon at unprecedented scale.

More like this →
arXiv Computer Vision
Sep 1

MANTLE: A Framework for Adaptive In-Situ Planetary Perception Using a Modular Uplink Principle

MANTLE is a multi‑task adaptive network designed for planetary perception, featuring a shared DINOv2 backbone with separate heads for landform classification and boulder segmentation. Trained on HiRISE and MSL imagery, it achieved 92.56% accuracy on seven Martian terrain classes and a 0.753 IoU for boulder segmentation, with strong cross‑sol generalization. The framework follows the Modular Uplink Principle, allowing lightweight task‑specific heads to be trained on Earth and uplinked to the rover without retraining the core model.

By Pranav Durai, Gary Doran
agentscomputer-visionrobotics
More like this →
arXiv Machine Learning
Oct 19

HiPredict: Predicting the missing HiRISE RED4 spectral channel with machine learning

HiPredict is a machine‑learning framework that reconstructs the failing RED4 spectral channel of the HiRISE instrument on the Mars Reconnaissance Orbiter using a U‑Net model trained on IR10, BG12, neighboring red‑channel data, and acquisition metadata. The model, trained on 81,717 samples and evaluated on 23,348 test samples, achieves a Normalised Mean Absolute Error of 1.1%, outperforming the existing synthetic red (SYN4) approach (3.1%). The reconstructed RED4 images enhance colour‑image recovery and better represent Martian surface features such as icy craters, ridge‑crest stratigraphy, and chloride deposits.

By Ardan Suphi, Ben Moseley, Gareth S Collins, Valentin T Bickel, Vidhya G Rangarajan
More like this →