Optimizing deep learning models for on-orbit deployment through neural architecture search
Open Access
- 29 October 2025
- journal article
- research article
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-13
Abstract
Advancements in spaceborne edge computing have facilitated the incorporation of Artificial Intelligence (AI)-powered chips into CubeSats, enabling intelligent data handling and enhanced analytical capabilities with greater operational autonomy. This class of satellites faces stringent energy and memory constraints, necessitating lightweight models typically obtained via compression techniques. This paper addresses model compression through Neural Architecture Search (NAS), enabling computational efficiency and balancing accuracy, size, and latency. More specifically, we design an evolutionary-based NAS framework for onboard processing and evaluate it on both burned-area segmentation and classification tasks. The proposed solution jointly optimizes network architecture and deployment for hardware-specific, resource-constrained platforms, with hardware awareness embedded in the optimization loop to tailor network topologies to the target edge computing chip. The resulting models, designed on CubeSat-class hardware-namely the NVIDIA Jetson AGX \(\text {Orin}^{\text {TM}}\) and Intel® \(\text {Movidius}^{\text {TM}}\) Myriad\(^{\text {TM}}\) X-exhibit a memory footprint below 1 MB, achieving real-time, high-resolution inference in orbit while outperforming handcrafted baselines in terms of latency (3\(\times\) faster) and maintaining competitive mean Intersection over Union (mIoU). Furthermore, on the classification benchmark conducted on an NVIDIA A100-SXM, our approach attained an atthew Correlation Coefficient (MCC) of 0.974-substantially outperforming the baseline EfficientNet-lite0 (0.902) while achieving a \(\times\)47 speedup. These results highlight the framework’s scalability, enabling seamless deployment across the spectrum from resource-constrained edge devices to datacenter-grade accelerators, thereby supporting next-generation on-orbit computing architectures.
Keywords
All Related Versions
This publication has 44 references indexed in Scilit:
- E2E: Onboard satellite real-time classification of thermal hotspots events on optical raw dataAstrodynamics, 2025
- Neural architecture search: A contemporary literature review for computer vision applicationsPattern Recognition, 2023
- EST-NAS: An evolutionary strategy with gradient descent for neural architecture searchApplied Soft Computing, 2023
- What is value in the New Space Economy? The end-users’ perspective on satellite data and solutionsActa Astronautica, 2023
- ViTAS: Vision Transformer Architecture Search2022
- Knowledge Distillation: A SurveyInternational Journal of Computer Vision, 2021
- A comprehensive survey on model compression and accelerationArtificial Intelligence Review, 2020
- Neural Architecture Search2019
- Cryosphere Applications2017
- A review of regional science applications of satellite remote sensing in urban settingsComputers, Environment and Urban Systems, 2013