The paper introduces a heterogeneous federated learning approach using the FractalNet architecture tailored for satellite mega‑constellations. It formalizes contact‑window‑constrained, depth‑heterogeneous optimization and proposes a distributed path scheduler that assigns model depth based on satellite SWAP‑C constraints, predicted contacts, and training statistics. The framework includes periodic update pooling and a three‑tier agentic control plane, and is validated through a wildfire detection case study across LEO, MEO, and GEO/HEO shells, demonstrating improvements in convergence, communication efficiency, energy adaptation, and robustness.
Satellite mega-constellations are emerging as large-scale sensing, communication, and computation fabrics, yet their learning architectures remain largely inherited from terrestrial federated learning...
arXiv:2608.24516v1 Announce Type: cross Abstract: Satellite-based distributed learning promises to train machine-learning models directly in orbit using massive, globally dispersed sensor data, there...
Satellite-based distributed learning promises to train machine-learning models directly in orbit using massive, globally dispersed sensor data, thereby avoiding large-scale data downloads to ground se...
arXiv:2608. 09687v1 Announce Type: new Abstract: Federated learning (FL) in Low Earth Orbit (LEO) satellite constellations is affected by non-IID data and irregular ground-station visibility, both driven by orbital geometry.
The paper introduces OrbitALIF, a federated learning framework that performs cloud removal on low‑earth‑orbit satellites. It uses a compact 2.30 M‑parameter spiking neural network with adaptive gated fusion and spectral‑spatial hybrid attention modules, enabling both training and inference onboard. The approach achieves competitive cloud‑removal quality while consuming only 0.287 mJ per inference on neuromorphic hardware, a 72.3‑fold energy reduction compared to an equivalent ANN.
arXiv:2608. 07126v1 Announce Type: cross Abstract: Most CubeSats, small and low-cost satellites roughly the size of a shoebox, do not survive as long as they were designed to: a study of 178 missions found that only 48-65% remain operational after two years, against a designed lifetime of 2-5 years.
arXiv:2607. 07758v1 Announce Type: new Abstract: Foundation models (FMs) have transformed machine learning from isolated task-specific model development toward general-purpose models pretrained on broad data and adapted to multiple downstream tasks.
The paper introduces a reinforcement‑learning‑guided evolutionary policy optimization framework for scheduling heterogeneous agile Earth observation satellites, addressing task selection, satellite assignment, and sequencing under diverse visibility windows, maneuvering constraints, energy use, and storage limits. It combines assignment‑based indirect encoding with decoder‑based cost evaluation to capture satellite‑dependent constraints while integrating task gain, energy savings, and load balance into a single utility metric. The resulting RLOSMEA algorithm uses reinforcement learning to select high‑level search operators, achieving higher weighted utility and more stable convergence than baseline metaheuristics across varied AEOS scenarios.
arXiv:2608. 09628v1 Announce Type: new Abstract: Collision avoidance systems are commonly used to avoid fragmentation events occurring in Low-Earth Orbit (LEO) and Geosynchronous Equatorial Orbit (GEO).
The paper introduces a reinforcement‑learning framework for managing bundles and flow control in deep‑space networks, where relay nodes must schedule and drop data under intermittent connectivity and limited storage. It defines a goal‑oriented loss based on excess age to prioritize mission‑relevant information and formulates relay control as a constrained semi‑Markov decision process solved with masked Lagrangian PPO. Simulations show that the approach improves age‑based utility and inter‑flow fairness compared to traditional schedulers.
The paper presents a formal framework for applying spiking neural networks to satellite onboard computing, deriving closed‑form energy models based on network sparsity, firing rate, and fan‑out. It introduces three novel spike‑based algorithms—SKCF for attitude estimation, TCAD for health monitoring, and SAMC for adaptive communication—alongside analytical complexity bounds. A quantitative Energy‑Accuracy‑Latency surface, populated with silicon data from Loihi and SpiNNaker, demonstrates that neuromorphic solutions can reduce energy consumption by 8 to 47 times compared to GPU baselines while keeping inference accuracy within 2.1% for typical satellite workloads.
arXiv:2608. 14789v1 Announce Type: new Abstract: Large low-Earth-orbit (LEO) Earth-observation (EO) constellations offer frequent access to geographically dispersed ground targets, but emergency requests may arrive after committed routine-plan execution has begun.