Papers for

satellite operators

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Intelligent satellites use orbital drift to save fuel observing targets

An intelligent satellite must do nothing well

Abstract: Autonomous inspection of a resident space object is usually posed as a contest between classical next-best-view planning and reinforcement learning. We argue that the choice of algorithm is the least informative of the design axes. An inspector spacecraft has no rest state: when it stops thrusting, orbital dynamics carry it along a natural trajectory that can be used to observe the target at no fuel cost. On a controlled ladder in which one design axis is varied at a time, fuel efficiency turns out to be set by three things: whether the agent is allowed to drift, whether its state records where drifting has taken it, and whether its planning horizon is long enough for the benefit to appear. Removing any one of them doubles the fuel bill or more. The algorithm still matters, but through what it can represent: a model-based agent whose learned world model ignores fuel spends several times more of it. Finally, no learned policy is safe by construction, and a control barrier function filter beneath the decision layer removed every keep-out-zone violation in two thousand evaluation episodes.

Mon 2 NovRobotics
The gist
Watching other spacecraft in space usually needs careful planning and uses a lot of fuel. The authors found that a satellite that lets itself drift naturally in orbit can observe a target using almost no extra fuel. How well this works depends on whether the satellite remembers where drifting took it and if it plans far enough ahead. They also showed that safety can be improved by adding a control filter to avoid forbidden zones automatically.
Open → 2611.02922v1

Energy aware split selection improves satellite cloud inference efficiency

STLT: A Novel System for Energy-Aware Distributed Inference in the Edge-Cloud-Space Continuum

Abstract: Low Earth Orbit satellites are pivotal for Earth observation tasks such as wildfire detection, capturing imagery for analysis by Deep Neural Networks (DNNs). On-board analysis reduces downlink volume, but severe energy constraints force a trade-off between computation and data transmission. Splitting the DNN between satellite and ground turns this balance into a design choice, where the split point determines combined computation and downlink energy. Existing approaches typically rely on static split-point decisions that fail to account for dynamic space conditions, or on complex algorithms, historical data, or prediction models that introduce significant overhead. To address these shortcomings, we present STLT, a two-stage system for energy-aware distributed inference between LEO satellites and a ground station. First, an offline analyzer characterizes the DNN architecture to identify a set of favorable split points, consequently reducing the search space. Second, an on-board algorithm performs lightweight energy-aware split selection, dynamically choosing from this pre-computed set based on the current state of the environment. We evaluate STLT on a real-world satellite telemetry dataset, showing that it reduces the split-candidate search space by 30.7% on average, lowers decision latency by 25.1% and energy by 27.5% on average, and delivers up to 20.1% more inference tasks than a state-of-the-art satellite-ground co-inference baseline while consuming 8.8% less energy overall.

Sat 31 OctDistributed, Parallel, and Cluster ComputingArtificial Intelligence
The gist
Satellites take pictures to spot things like wildfires but have limited energy to process and send data. The authors designed a system called STLT to smartly decide where to split the work between the satellite and ground to save energy. It uses a two-step method that first finds good split points and then picks the best one on the fly based on current conditions. This approach lets satellites analyze more images while using less energy compared to older methods.
Open → 2611.01233v1

On-orbit servicing mission planning includes low-thrust propulsion dynamics

Integrating Low-Thrust Transfer Dynamics into POMDP Planning for On-Orbit Servicing

Abstract: Recent developments in both high-thrust (HT) and low-thrust (LT) propulsion technologies broaden design options for on-orbit servicing (OOS) missions. However, incorporating LT transfers into OOS mission planning requires accounting for the trade-off between prescribed transfer duration and propellant consumption. Multi-servicing, multi-transfer OOS mission planning is inherently sequential and requires adaptive planning tools that account for propulsion-dependent transfer durations and mass-dependent propellant consumption. Thus, transfer dynamics must be integrated into mission planning and scheduling to increase flexibility and adaptability. In addition, the planning is often done with imperfect information about the system's state and stochastic service demands. This paper focuses on modeling orbital transfers, particularly LT transfers, within a Partially Observable Markov Decision Process (POMDP) framework for OOS mission planning. To enable offline policy computation, the transfer model must provide explicit state transitions and rewards for all admissible actions. The state space is therefore augmented with the servicing vehicle's remaining propellant to enable consistent mass updates. HT transfers are modeled using a precomputed transfer matrix that maps each transfer to the time of flight and propellant consumed. LT transfers are represented using an analytical GEO phasing model with prescribed transfer-duration alternatives. For each alternative, the model calculates propellant consumption and maneuver feasibility as functions of the required phase change and the current servicing-vehicle mass. The proposed modeling approach is evaluated on simplified OOS planning scenarios using circular phasing with HT and LT propulsion. The resulting POMDPs are solved using the Successive Approximations of the Reachable Space under Optimal Policies (SARSOP) algorithm, and the policies obtained under HT and LT are compared. This comparison provides a quantitative assessment of the feasibility and operational impact of LT options, particularly regarding the potential role of advanced electric propulsion in future servicing architectures.

Fri 30 OctRobotics
The gist
Planning space missions to service satellites requires deciding how to move the servicing vehicle between targets using different engines. Low-thrust engines save fuel but take longer, so planning must balance time and propellant use. The authors show how to include these trade-offs in a decision-making model that handles uncertainty and changing conditions. They use mathematical tools to generate mission plans that consider both high- and low-thrust options and compare their impacts. This helps future space servicing missions use electric propulsion effectively.
Open → 2610.41839v1

Autonomous spacecraft inspection improves safety with chance constraints

Chance-Constrained Inspection Viewpoint Sequencing for On-Orbit Servicing via Safe Reinforcement Learning

Abstract: Autonomous inspection of a target spacecraft is typically the initial step in on-orbit servicing. Planning close-proximity observation viewpoints involves a complex combinatorial problem constrained by coverage, fuel, and safety considerations. This paper models the viewpoint selection as a Constrained Markov Decision Process (CMDP) and applies Proximal Policy Optimization with Lagrangian relaxation to solve it. The system combines a high-level planner, which sequences camera viewpoints on an observation sphere around the target, with a low-level model that assesses each transfer using the Clohessy--Wiltshire (CW) equations for relative motion. The deterministic constraints on single-impulse reachability and $Δv$ cap are enforced directly on the action set. Safety margins are affected by navigation and manoeuvre errors, making the actual clearance stochastic. To address this, a chance constraint on the realised trajectory is implemented, with a Lagrangian multiplier governed by PID control to adapt the penalty during training. The approach is tested on a Galileo-class navigation satellite in medium Earth orbit, modelled as a point cloud where visibility from each viewpoint considers field-of-view and line-of-sight occlusion. Across various risk budgets, the trained policy maintains the keep-out violation rate within the specified limit, achieving coverage close to its maximum and staying within the mission $Δv$ budget. Compared to the usual practice of applying a fixed margin to the keep-out radius and planning deterministically, the chance-constrained policy attains less than half the violation rate at comparable coverage. This makes the collision risk an explicit design parameter, eliminating the need for re-tuning per case. The learned policy selects a viewpoint without the repeated Monte Carlo risk estimates required by a greedy, chance-constrained planner at every decision point.

Fri 30 OctArtificial Intelligence
The gist
Planning safe close-up inspections of spacecraft in orbit is tricky because you have to cover the whole target while saving fuel and avoiding collisions. The authors treat this task as a decision problem with safety limits and use a type of machine learning to find the best sequence of camera positions. Their method adapts the plans based on the risk of errors in navigation and movements, making safety violations rare and predictable. Tests showed it works better than standard methods that use fixed safety buffers, balancing coverage, fuel use, and collision risk.
Open → 2610.41639v1

Machine learning advances modeling and forecasting of space weather impact

Machine learning applications in the heliosphere

Abstract: Space weather is the study of the near-Earth impact of the processes on the Sun. In a this era of technology, space weather emerges as a crucial area of study wherein the state of the near-Earth conditions need to be modeled and forecasted. Crucially, the advent of machine learning provides us with opportunities to study this subject in a rather comprehensive manner. In this chapter, we treat the the processes arising from Sun to the impact on the ground as a set of coupled systems, which can be studied in isolation or in a coupled manner. We define these individual systems in terms of their state vector $\vec{S}$, which provides us with an opportunity to define the physical variables determining the system. This also allows us to consider the systems in terms of their observables, obtained through the action of operators as $\vec{O}[\vec{S}]$. In such a formalism, the machine learning algorithms serve as {\it learned operators}, defined generally as inference ($\mathcal{I}$), propagation ($\mathcal{P}$), and emulation ($\mathcal{E}$). Through such a formalism, we present how machine learning helps advance our understanding of the physical systems under study, while also accelerating the development of operational models for space weather forecasting.

Tue 27 OctMachine Learning
The gist
Space weather describes how the Sun’s activity affects Earth, which is important for technology and safety. The authors treat the Sun-to-Earth system as linked parts that can be studied separately or together using specific physical measurements. Machine learning is used to learn how these parts behave, helping to predict space weather more accurately and quickly. This approach not only improves scientific understanding but also helps build practical forecasting tools.
Open → 2610.37927v1

Topology of solar images helps predict high-speed solar wind streams

Explainable Prediction of High-Speed Solar Wind Streams driven by Coronal Holes via Topological Data Analysis

Abstract: High-speed solar wind streams (HSSs) originating from coronal holes (CHs) are an important driver of recurrent geomagnetic activity. Since HSSs are associated to coronal structures visible in extreme-ultraviolet observations, this provides a natural basis for forecasting several days in advance. In this work, we investigate whether the topology of coronal structures observed in extreme ultraviolet (EUV) images contains predictive information on the subsequent occurrence of CH-driven HSSs at Earth. To this end, we consider Topological Data Analysis (TDA), and in particular persistent homology, to classify EUV solar images according to the occurrence of HSSs at Earth four days later. The target is a binary classification of HSS occurrence at a four-day lead time, defined using an HSS interval catalog and solar wind speed thresholds. Several combinations of topological representations and classifiers are evaluated under a common cross-validation framework. The results indicate that the topology of EUV coronal structures alone contains predictive information on subsequent CH-driven HSS occurrence, despite the absence of explicit CH geometric descriptors in the model input. To investigate the origin of this predictive information, we develop a model-level explainability framework based on cross-validation-aggregated feature weights and spatial evidence maps. A fully traced example connects persistence diagrams and persistence-image features to the classifier output and to their spatial localization in the original EUV observation. The proposed framework provides a transparent and computationally lightweight approach for investigating the relationship between coronal topology and subsequent high-speed solar wind streams.

Mon 26 OctMachine Learning
The gist
High-speed solar wind streams from the Sun can affect Earth’s magnetic environment. These streams come from dark areas called coronal holes that show up in special solar images. The authors used a method called Topological Data Analysis (TDA) to study the shape and connectivity of these dark areas to see if they could predict when the solar wind streams would arrive at Earth four days later. Their results show that just the topological features of the images carry useful information for forecasting these solar wind events. They also created a way to explain how the topological data links to the predictions, making their approach more understandable.
Open → 2610.35549v1

Satellite runtime enables continuous app migration in low Earth orbit

Migration as a Steady-State Execution Primitive: A Runtime for Virtual Stationarity in Low-Earth Orbit

Abstract: Low Earth Orbit (LEO) satellite mega-constellations enable orbital edge computing for latency-critical geocentric applications, but satellite motion compresses line-of-sight communication windows to mere minutes. To maintain logical geographic proximity, a paradigm termed Virtual Stationarity (VS), applications must continuously migrate across passing satellites in the constellation. Traditional live migration mechanisms optimize individually initiated relocations or replication, but not the continuous handover between successive hosts that orbital motion requires. We reformulate live migration as a steady-state execution model and introduce a novel virtual stationarity runtime paradigm. We formalize the viability of continuous migration through a lightweight model capturing latency and availability service-level objectives (SLOs), continuous migration overhead, and the convergence bound of state replay. To satisfy these constraints, we derive a Virtual Stationarity Runtime (VSR) framework whose Migration Membrane combines Capability Virtualization and Continuous State Projection to rebind host resources and reconstruct evolving execution on a successor. Evaluation on physical heterogeneous hosts shows 95th-percentile downtime below 6 milliseconds across 300 recurrent handovers, migration time effectively invariant to memory-write rate, and no cumulative growth in migration time or continuation state. A 1,600-satellite simulation further confirms completion within orbital handover windows.

Fri 23 OctDistributed, Parallel, and Cluster ComputingNetworking and Internet Architecture
The gist
Satellites in low Earth orbit move quickly, making it hard for applications to stay near a specific geographic location in space. The authors propose a new way to keep apps running seamlessly by constantly moving them from one satellite to the next as the satellites pass by. Their system, called the Virtual Stationarity Runtime, handles this continuous migration efficiently with very little interruption. Tests show this approach keeps downtime low and works well even with many satellites involved.
Open → 2610.32720v1

Spacecraft controls adjust speed to improve attitude accuracy safely

Paired-Outcome Learning for Selective Spacecraft Attitude Capture under Torque Limits

Abstract: Torque-limited spacecraft attitude capture requires adapting the maneuver speed to uncertain braking authority. Faster settings can reduce average tracking error while worsening individual maneuvers, making selective modification of an existing controller a baseline-relative learning problem. A paired-outcome method combines three mechanisms: physically interpretable continuation policies with different speed envelopes, separate prediction of task failure and successful-pair tracking gains, and context-dependent rejection that executes the unchanged baseline. Complete paired simulations provide labels under common plant parameters and measurement seeds; training, validation, and calibration are separated by context. Basilisk experiments use 108 development contexts, 324 labeling rollouts, and 624 evaluation episodes across 48 test contexts, including disclosed post-freeze diagnostics. The selective controller reduces aggregate integrated attitude error by 4.67% relative to the branch baseline, with 17 improvements, 31 exact ties, and no observed regressions. Always using the fast alternative achieves a 4.81% reduction but worsens 11 cases. Gains occur in principal-axis, near-axis, braking-boundary, and selected noise/drift maneuvers; general coupled cases retain the baseline. Ablations show no additional aggregate benefit from branch commitment, active querying, or score subtraction. The evidence supports a restricted empirical trade-off between average improvement and individual regression, without establishing a closed-loop safety guarantee.

Sat 17 OctMachine Learning
The gist
Controlling the orientation of spacecraft is tricky when the amount of force available to slow them down is uncertain. The researchers created a way to selectively choose between slow and fast control settings based on the situation, aiming to reduce errors without making any maneuver worse. They tested their method extensively in simulations and experiments and showed it improves overall accuracy while avoiding bad outcomes. Their approach uses paired examples to learn when to stick with the original control or switch to a faster option.
Open → 2610.24763v1

AI detects ships onboard satellites to cut maritime data sent earthward

AI-Based On-Board Maritime Object Detection for Earth Observation Payload Data Reduction on Versal Embedded Hardware

Abstract: Very-high-resolution Earth-observation satellites acquire more data than they can store and downlink, while in maritime surveillance the vessels cover a tiny fraction of each scene. We study onboard vessel detection as a way to select what is downlinked, which reduces the data according to its content rather than coding every pixel; it is complementary to conventional onboard compression. The work follows three axes. (i) Data and algorithm: a controlled dataset is generated from 68 annotated Maxar scenes with 43 vessel classes, and a YOLOX-S detector is trained on it. (ii) Embedded deployment: the detector is quantized and deployed on the DPU of a Versal VC1902, with a limited loss of detection quality and a processing time of a few seconds per scene. (iii) Data reduction: we propose several downlink modes, from metadata only (box, class and score of each detection) to image crops around vessels, tiles holding detections, or the whole scene with a degraded background, and estimate from the measured detection errors the trade-off each offers between the vessels kept and the volume downlinked. On our dense harbor and coastal scenes, tiles keep 98% of the vessels with 29% of the scene volume, and crops 83% with 3%.

Thu 8 OctComputer Vision and Pattern RecognitionArtificial Intelligence
The gist
Earth observation satellites collect huge amounts of data, but most of it does not contain ships, which are of interest for maritime surveillance. The authors created a new dataset and trained an AI model to spot vessels directly onboard the satellite. They deployed this AI on embedded hardware, so it can quickly process scenes and decide which parts contain ships worth sending to Earth. This approach cuts down data transmission significantly while keeping almost all ships visible in the images.
Open → 2610.12182v1

Onboard detection of unusual marine activity improves space observation

Onboard Marine Anomaly Detection on $Φ$sat-2: From Simulation-Based Development to In-Orbit Demonstration

Abstract: Onboard Artificial Intelligence can improve responsiveness and bandwidth efficiency of Earth Observation systems by processing data directly on the satellite. This paper presents the experience gained from the development, onboard integration, and post-launch adaptation of a lightweight marine anomaly detection pipeline deployed on the European Space Agency's $Φ$sat-2 mission. The application combines sea segmentation, self-supervised feature encoding of marine regions, generic anomaly detection based on deviations from a normal sea state, and optional characterization of selected anomaly types. Before launch, the pipeline was trained and validated on simulated $Φ$sat-2 imagery to assess algorithmic performance and compatibility with resource-constrained onboard hardware. After integration and functional validation in the mission environment, early experiments on real $Φ$sat-2 acquisitions revealed a significant mismatch between simulated and in-orbit data. The pipeline was therefore retrained on real Level-1 imagery using an improved annotation strategy to better handle ambiguous marine regions, substantially enhancing performance. Beyond demonstrating the onboard feasibility of the application, the $Φ$sat-2 experience highlights the importance of robust annotation strategies and sensor-aware design, and shows that simulation-based development is valuable for pre-flight risk reduction, while reliable scientific validation requires representative in-orbit data and should be clearly distinguished from functional validation.

Thu 8 OctArtificial IntelligenceComputer Vision and Pattern Recognition
The gist
Satellites that watch the Earth can work better if they can spot unusual things happening in the ocean themselves, without sending everything back to Earth first. The authors describe how they built and tested a lightweight system on a European satellite to detect strange marine events like unusual water patterns. Before launch, they trained their system on simulated satellite images, but after launch, they found the real images were different and had to retrain the system with real data to improve accuracy. This work shows that building space AI needs careful preparation with real data and smart design to work well in orbit.
Open → 2610.11735v1

Satellite system spots unusual marine events for better monitoring

On-Board Anomaly Detection for Efficient Marine Environmental Monitoring

Abstract: Marine ecosystems are impacted by various threats such as oil spills, algal blooms, and sediment floods, which disrupt habitats, wildlife, and human activities. Advances in satellite imagery and Artificial Intelligence (AI) have enhanced our capabilities for early detection and mitigation of such hazards. In this paper, we propose a marine event detection pipeline for Earth observation satellites equipped with multi- or hyperspectral sensors. Our approach includes a self-supervised neural network encoder that compresses satellite images into a reduced latent space, enabling efficient onboard processing. A machine learning anomaly detection model identifies deviations from normal sea patterns to detect environmental anomalies. We compare its performance against traditional algorithms such as Isolation Forest, One-Class Support Vector Machine and Local Outlier Factors. Our lightweight, resource-efficient pipeline is optimized for deployment on satellites with limited computational resources, ranging from embedded CPUs to AI hardware accelerators. By prioritizing the transmission of critical information, our solution enhances system responsiveness and optimizes satellite communication bandwidth. Demonstrated through current integration across multiple missions, including European Space Agency's (ESA) Phisat-2 mission and Microsoft/Thales Alenia Space IMAGIN-e mission, our pipeline aims to improve marine environmental monitoring by providing timely alerts and efficient data reduction.

Fri 2 OctComputer Vision and Pattern RecognitionArtificial IntelligenceMachine Learning
The gist
Oceans face problems like oil spills and harmful algae that disturb sea life and people’s activities. The authors offer a way to detect unusual ocean conditions by analyzing satellite images directly onboard, using smart AI to spot odd patterns without sending all data back to Earth. Their method compresses the images and flags anomalies more efficiently than older techniques. This helps satellites send alerts faster and saves communication resources.
Open → 2610.03649v1

Spacecraft swarms coordinate power use with magnetic controls

Decentralized Power-Optimal Coordination for Spacecraft Swarms Using Time-Varying Magnetorquer Actuation

Abstract: This paper presents a decentralized power-optimal coordination framework for magnetically actuated spacecraft swarms. Swarms that form large space structures overcome the aperture limit set by the launch vehicle and hold their shape on solar-generated power alone. Magnetic actuation is propellant-free and generated by a magnetorquer, which is commonly used for attitude control. However, every spacecraft interacts with every other within range, and its effect depends on the actuation power and a carrier frequency. We therefore design a decentralized power-optimal framework to jointly derive the interaction graph, frequency grouping, and controller gains. Our decentralized controller preserves angular momentum, which is a nonholonomic constraint. Then, this framework for connected groups whose memberships overlap across carriers guarantees that the relative position errors, the absolute attitude errors, and the imbalance of the reaction-wheel momenta converge to the desired states under the decentralized power-optimal allocation. A closed-loop simulation of a thousand spacecraft with the complete alternating-current interaction confirms the framework. A fast approximate integration with a proven error bound extends the framework to a long-horizon orbital reconfiguration held with high precision.

Thu 1 OctMultiagent Systems
The gist
Spacecraft can work together in swarms to form large structures in space, but coordinating their movements precisely while saving energy is challenging. The authors present a method that lets each spacecraft control itself and communicate indirectly with others using magnets, without needing fuel. Their approach finds the best way to share power while keeping the whole group stable and in the right formation. They tested this with simulations involving a thousand spacecraft, showing it works well even for complex, long missions.
Open → 2610.02118v1

Flow matching improves speed and detail in SAR to optical image conversion

ContraFM-S2O: Flow Matching-Based One-step SAR-to-Optical Image Translation Model with Contrastive Learning

Abstract: In recent years, diffusion models and GAN-based models have become the mainstream approaches for SAR-to-optical image translation, owing to their advantages, such as high-quality generation and stable training. However, they have shortcomings such as high inference latency and the generated optical images suffer from low detail fidelity, often resulting in blurred edges and loss of fine textures. Thus, we propose ContraFM-S2O, which is a flow matching-based model for SAR-to-optical image translation. Unlike conventional diffusion models, ContraFM-S2O learns to predict the velocity field in training and solves ODE instead of SDE during inference to improve the sampling efficiency. In addition, ContraFM-S2O replaces instantaneous velocity with average velocity along the interpolation path to realize one-step SAR-to-optical image translation and uses contrastive learning to improve the quality of the generated optical images. Experiments show our model achieves state-of-the-art on SAR2Opt and QXS datasets, outperforming baselines, and reduces inference latency via one-step generation.

Fri 25 SeptComputer Vision and Pattern Recognition
The gist
Turning radar images (SAR) into regular photos that people can understand is hard because details often get blurry or lost in the process. The authors created a new method called ContraFM-S2O that uses a mathematical technique called flow matching to convert images faster and with clearer details. Instead of taking many slow steps, their approach predicts the full transformation in one go. They also use a learning method called contrastive learning to make the converted images more accurate and sharp. Tests show their method works better and faster than older methods on common datasets.
Open → 2609.31378v1

Physics informed ai detects solar wind space weather events early

PISCES: Physics-Informed Solar-wind Convolutional autoEncoder for Space-weather Anomaly Detection and Early Warning

Abstract: Space weather early warning depends on detecting solar wind transients in in-situ measurements at the first Sun-Earth Lagrange point (L1), before they reach Earth. Fixed thresholds can miss combined magnetic and plasma structure, and many learning methods provide a single anomaly score. We present the Physics-Informed Solar-wind Convolutional autoEncoder for Space-weather (PISCES), a convolutional autoencoder trained without catalog labels on OMNI solar wind measurements under physics constraints. Its loss includes magnetic field consistency, an empirical relation between temperature and velocity, the Parker spiral angle, and penalties on changes between consecutive one-minute samples in derived quantities calculated from the reconstruction. At inference, PISCES separates the anomaly score into magnetic and plasma reconstruction errors, physics relations, and residual corrections, and reports the magnitude of each contribution. Attenuation of the skip connections, selected on validation data, improves average precision for the trained models, while the untrained scores remain nearly the same. The trained models also give a more consistent ordering of these physical contributions. After smoothing with a trailing median, the alarms can precede independently observed sudden commencements, including positive sudden impulses.

Wed 23 SeptMachine LearningArtificial Intelligence
The gist
Space weather events caused by bursts of solar wind can affect Earth, so detecting these early is important. The authors developed an AI model called PISCES that learns from measurements of the solar wind while respecting known physics relationships. Instead of one generic alert, it reports detailed scores showing what kind of changes occurred in magnetic fields and plasma. This detailed early warning can help identify sudden events that might impact Earth before they arrive.
Open → 2609.28022v1

Vision foundation model improves spacecraft pose estimate accuracy

Vision Foundation Models with Synthetic-Only Training for Monocular Spacecraft Pose Estimation

Abstract: We present an improvement on previous spacecraft pose estimation architectures that results in the lowest published mean rotation errors we know of on the SPEED+ lightbox and sunlamp test sets for a known, non-cooperative spacecraft. By using a previously established heatmap-based pose estimation architecture and adapting a large self-supervised ViT foundation model (DINOv3) in place of the smaller convolutional and ViT encoders of previous work, we show that pose estimation accuracy improves from 300M to 840M parameters with no saturation yet observed. We also evaluate our 840M model on a Jetson Orin NX 16GB, measuring single-pass network inference at 133.8 ms per crop with a board draw of 32.0 W. These measurements demonstrate embedded inference feasibility on a processor family with orbital flight heritage. Our resulting model outperforms previous models across lightbox and sunlamp domains while training only on synthetic data. Our best model, using DINOv3 840M adapted with LoRA as the encoder (rank 64, three-seed ensemble with four-rotation test-time augmentation), results in $1.56^\circ$ mean rotation error on sunlamp and $1.17^\circ$ on lightbox, compared to the previous best mean rotation errors we know of on these test sets, $2.66^\circ$ and $1.75^\circ$ by EagerNet.

Tue 22 SeptComputer Vision and Pattern RecognitionRobotics
The gist
Estimating the position and orientation of spacecraft from a single camera image is important but challenging. The authors improved this task by using a much larger and more advanced vision model than before, which was trained only on simulated images. Their approach achieved more accurate results than previous methods on standard test environments that mimic real spacecraft lighting conditions. This shows that powerful models trained on just synthetic data can work well for important space applications.
Open → 2609.26561v1

Satellite swarms use distributed computing to cut communication needs

A Distributed Computing Framework for Satellite Swarms

Abstract: The rise of large satellite constellations and Distributed Space Systems (DSS) demands generalized frameworks that enable fault-tolerant, autonomous distributed space applications. Conventional ground-centric command and control does not scale to systems of tens or hundreds of satellites, motivating the adoption of distributed computing. This paper introduces a conceptual distributed computing framework for satellite swarms, covering distributed state, command and control, and scientific mission. As a first validation step, a strongly eventually consistent distributed state service is designed and implemented using Conflict-free Replicated Data Types (CRDT), specifically a Last-Write-Wins Register-based key-value store. The service is evaluated in the context of Space Situational Awareness catalog dissemination across a simulated 66-satellite constellation emulated with GoNetEm. Results show that the CRDT-based approach reduces ground-to-satellite communication from 66 messages to a single uplink per update, with tree traversal matching the total message count of direct uplink. Under concurrent updates, sub-linear inter-satellite message growth is observed as nodes discard outdated versions, further reducing network overhead. These results demonstrate the viability of a CRDT-based distributed state as a scalable and fault-tolerant foundation for satellite swarm applications.

Wed 16 SeptDistributed, Parallel, and Cluster ComputingNetworking and Internet Architecture
The gist
Managing large groups of satellites is hard because each one needs lots of instructions from Earth. The paper shows how satellites can share information among themselves using a system called CRDTs, which helps keep data consistent even with delays or errors. This reduces the number of messages sent from the ground a lot. The authors tested their idea on a simulated group of 66 satellites and found it cuts communications and handles conflicts well.
Open → 2609.18839v1

Autonomous spacecraft maneuver planning improves collision avoidance under uncertainty

Chance-Constrained Belief-Space Maneuver Planning for Autonomous Collision Avoidance Under Uncertainty

Abstract: Increasing conjunction frequency in low Earth orbit places growing pressure on spacecraft operators to determine not only whether an encounter requires mitigation, but whether sufficient information is available to commit to a maneuver. This work formulates this information-action tradeoff as a belief-space planning problem for conjunctions between a maneuverable spacecraft and an unmaneuverable secondary object. The planner represents the uncertain orbital states as Gaussian beliefs and uses a chance-constrained belief-space Monte Carlo tree search framework to reason over possible future tracking updates before time of closest approach (TCA). A terminal chance constraint limits the probability of reaching TCA above a prescribed collision-risk threshold, allowing the planner to wait for informative tracking while intervening when deferral becomes too risky. We evaluate the approach on eight historical conjunctions from NASA's Conjunction Assessment Risk Analysis dataset. By varying the secondary-object measurement quality and tracking cadence, we generate a total of 96 distinct evaluation scenarios. Across the evaluated conditions, the planner reaches TCA without maneuvering in approximately 40% of episodes while maintaining no terminal collision-risk violations. In contrast, fixed-time rule-based maneuver policies resolve more encounters without maneuvering when intervention is deferred closer to TCA, but at the expense of increasing terminal risk violations. The fraction of episodes reaching TCA without maneuvering depends strongly on tracking quality and measurement cadence, ranging from 76% under accurate, frequent measurements to approximately 18%-20% under the poorest tracking conditions. These results show that tracking quality and frequency are not only inputs to collision-risk estimation: they can determine when intervention becomes necessary.

Fri 11 SeptRoboticsArtificial Intelligence
The gist
Spacecraft operating in low Earth orbit face more frequent chances of colliding with objects. The authors created a planning method that helps decide when a spacecraft should move to avoid collisions based on how much is known about the uncertain positions of nearby objects. Their method waits for better tracking data when possible, but maneuvers if the risk becomes too high. Tests on real historical data showed this approach could safely reduce unnecessary maneuvers while maintaining collision safety.
Open → 2609.13428v1

Deep learning improves faint space object detection by removing stars

Improving Faint Object Detection for Space Situational Awareness with Variational Autoencoders

Abstract: We present a deep-learning pipeline for enhancing the detection of faint moving objects in optical space situational awareness (SSA) imagery through automated star removal and background reconstruction. Detecting low signal-to-noise ratio (SNR) objects remains extremely challenging in optical observations, particularly in the cislunar (X-GEO) environment, where structured sky backgrounds, dense stellar fields, and scattered moonlight significantly degrade the performance of classical detection algorithms. To address this problem, the proposed pipeline combines a lightweight segmentation network (Tiny-U-Net) to generate stellar masks with a partial-convolution variational autoencoder (astro-VAE), designed to learn the statistical distribution of astronomical backgrounds and perform context-aware inpainting of masked regions. The reconstructed background maps can then be used as a preprocessing step to suppress fixed sources and background inhomogeneities prior to detection. As a proof of concept, the approach is integrated with a shift-and-stack scheme and evaluated on real ground-based telescope observations targeting the X-GEO region. Results demonstrate that the method reconstructs star-free backgrounds with high fidelity, while preserving moving targets and significantly enhancing detectability, thereby providing an effective data-driven preprocessing strategy for faint moving-object detection in optical SSA scenarios.

Thu 10 SeptComputer Vision and Pattern RecognitionArtificial IntelligenceMachine Learning
The gist
Detecting very faint moving objects in space images is hard because bright stars and background light make them hard to see. The authors created a deep-learning method that first identifies and masks stars in telescope images, then uses a special autoencoder to fill in the missing areas realistically. This process cleans up the background, making it easier to spot faint moving objects like satellites or debris. When tested on real telescope data, the method improved the detection of these faint objects without erasing them.
Open → 2609.11269v1

Maneuver annotated dataset improves tracking of low Earth orbit satellites

MAD-LEO: A Maneuver-Annotated Orbital Dataset for LEO Satellites with Tiered Multi-Source Evidence

Abstract: With the rapid development of aerospace technology and the large-scale deployment of low Earth orbit (LEO) constellations, the risk of orbital collisions has increased, creating a growing demand for reliable observations of satellite maneuvers. However, public datasets containing real maneuver records remain scarce. We present MAD-LEO, a Maneuver-Annotated orbital Dataset for LEO satellites. The mission-reported subset contains 1,134 maneuver events from eleven geodetic and altimetry satellites spanning 1992 to 2026, with labels taken directly from mission-published maneuver histories. Each event is checked against two-line element (TLE) data, precise orbit products, and satellite laser ranging (SLR) observations, with evidence tiers assigned according to data availability. The operational subset pairs operator-published ephemerides for 6,785 Starlink satellites with cataloged TLE records over a continuous 107-hour period. Technical validation across seven machine-readable experiment suites confirms the cross-source consistency of the labels and the evidence products.

Tue 8 SeptDatabases
The gist
As many satellites orbit closer to Earth, avoiding crashes becomes harder and more important. The authors created MAD-LEO, a dataset that records when satellites change their path (maneuvers) using reliable sources like mission reports, orbital data, and laser measurements. This dataset helps confirm and cross-check satellite movements, which are usually hard to track and find in public records. They tested MAD-LEO to ensure the data is consistent and useful for monitoring satellites.
Open → 2609.08556v1