First On-Sky Demonstration of Reinforcement Learning for Adaptive Optics Control
Researchers have demonstrated the first on-sky use of a reinforcement learning algorithm, called PO4AO, to control an adaptive optics system on a real telescope. The system was deployed on the Papyrus AO system at the 1.52 m telescope at the Observatoire de Haute-Provence and consistently outperformed the standard integrator controller across varying atmospheric conditions and flux levels. The result marks a significant step toward broader adoption of machine learning-based control in astronomical instrumentation.
A team of researchers has reported the first on-sky validation of a reinforcement learning (RL)-based adaptive optics controller, named Policy Optimization for AO (PO4AO), accepted for publication in Astronomy & Astrophysics. The system was deployed on the Papyrus adaptive optics instrument at the 1.52 m telescope at the Observatoire de Haute-Provence (OHP) in France, interfaced with the existing real-time controller via shared-memory buffers using a Python implementation. Across multiple nights of observation spanning a range of atmospheric seeing conditions and stellar flux levels, PO4AO consistently outperformed the conventional integrator controller used as a baseline. The algorithm demonstrated the ability to learn and compensate for telescope vibration patterns and showed strong robustness to measurement noise. Notably, once tuned for the Papyrus system, PO4AO operated in a turnkey fashion with a single set of hyperparameters, requiring no per-target or per-condition adjustment. These gains were achieved despite a non-optimized Python implementation that introduced approximately 750 microseconds of additional latency, along with control jitter and occasional frame drops, suggesting further performance improvements are possible with a more optimized deployment. The authors conclude that PO4AO represents a robust, high-performance controller for single-conjugate adaptive optics and paves the way for RL strategies in operational astronomical observatories.
What's missing
The generalizability of PO4AO to larger-aperture telescopes or more complex multi-conjugate AO systems remains an open question, as the demonstration was limited to a single 1.52 m telescope. The computational requirements and training time needed before deployment on a new system are not fully characterized.
What different sources said
- arXiv cs.LGCenter
On-sky demonstration of reinforcement learning for adaptive optics control
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.