AutoRASOR: Autonomous Rapid Scanning Electron Microscope Operator
Abstract
Scanning Electron Microscopy (SEM) is a foundational technique for characterizing material microstructure, which dictates many fundamental physical and chemical properties. With the rise of Self-Driving Labs (SDLs), samples are now synthesized in large batches, demanding equally high-throughput, autonomous characterization. Existing automated electron microscopy pipelines are task-specific: they detect pre-defined features, optimize known properties, or require prior knowledge of the sample, restricting each pipeline to the material system it was built for. We introduce AutoRASOR, a task-agnostic autonomous SEM pipeline that captures the multi-scale morphology of an unknown sample without domain-specific pre-training, fine-tuning, or human prompting. AutoRASOR embeds micrographs in real time with a vision foundation model (VFM), DINOv3, and selects regions of interest (ROIs) through two complementary policies: Latent Farthest Point Sampling (LFPS) and active learning on morphological ambiguity, which is the conditional variance of unresolved fine-scale features given lower magnification appearance. Active learning on ambiguity consistently captures more diverse and rare morphologies than random ROI selection, while LFPS reliably recovers the specimen's morphological distribution within a limited capture budget, tested both on real SEM micrographs and synthetic phase-field images. By shifting autonomous characterization from pre-defined, task-specific targeting of features to morphological survey, AutoRASOR generates information-rich multi-scale datasets for diverse downstream tasks. Without the need for domain-specific pre-training or prior material assumptions, this framework can be deployed out-of-the-box to characterize novel materials produced by SDLs.
Related
- Attosecond current control and timing in a scanning tunnelling microscope
- Modified Low-Temperature Scanning Tunneling Microscope for Ultrafast Pump-Probe Spectroscopy
- Probing (sub)nanoscale ferrons in an electron microscope
- Extending the operating window of scanning electron microscopy through an integrated electron-optical architecture for high-temperature and near-ambient-pressure environments
- Multi-Angle Beam Scanning Transmission Electron Microscopy for Diffraction-Contrast-Free Magnetic Imaging
BibTeX
@misc{zhang2026autorasor,
title = {{AutoRASOR: Autonomous Rapid Scanning Electron Microscope Operator}},
author = {Kevin Zhang and Mohammad Taha and Rafael Espinosa Castañeda and Yutong Liu and Yin Zhu and Lotan Portal and Robert A. Mcleod and Vahid Attari and Jane Y. Howe and Jason Hattrick-Simpers},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.22411},
eprint = {2609.22411},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.22411},
}
RIS
TY - UNPB
TI - AutoRASOR: Autonomous Rapid Scanning Electron Microscope Operator
AU - Kevin Zhang
AU - Mohammad Taha
AU - Rafael Espinosa Castañeda
AU - Yutong Liu
AU - Yin Zhu
AU - Lotan Portal
AU - Robert A. Mcleod
AU - Vahid Attari
AU - Jane Y. Howe
AU - Jason Hattrick-Simpers
PY - 2026
DA - 2026/09/18
PB - arXiv
DO - 10.48550/arxiv.2609.22411
UR - https://arxiv.org/abs/2609.22411
AB - Scanning Electron Microscopy (SEM) is a foundational technique for characterizing material microstructure, which dictates many fundamental physical and chemical properties. With the rise of Self-Driving Labs (SDLs), samples are now synthesized in large batches, demanding equally high-throughput, autonomous characterization. Existing automated electron microscopy pipelines are task-specific: they detect pre-defined features, optimize known properties, or require prior knowledge of the sample, restricting each pipeline to the material system it was built for. We introduce AutoRASOR, a task-agnostic autonomous SEM pipeline that captures the multi-scale morphology of an unknown sample without domain-specific pre-training, fine-tuning, or human prompting. AutoRASOR embeds micrographs in real time with a vision foundation model (VFM), DINOv3, and selects regions of interest (ROIs) through two complementary policies: Latent Farthest Point Sampling (LFPS) and active learning on morphological ambiguity, which is the conditional variance of unresolved fine-scale features given lower magnification appearance. Active learning on ambiguity consistently captures more diverse and rare morphologies than random ROI selection, while LFPS reliably recovers the specimen's morphological distribution within a limited capture budget, tested both on real SEM micrographs and synthetic phase-field images. By shifting autonomous characterization from pre-defined, task-specific targeting of features to morphological survey, AutoRASOR generates information-rich multi-scale datasets for diverse downstream tasks. Without the need for domain-specific pre-training or prior material assumptions, this framework can be deployed out-of-the-box to characterize novel materials produced by SDLs.
ER -