Satellite edge computing solutions are promising to usher in a new era of efficiency and versatility onboard spacecraft of all sizes.
In this article we take a closer look at how edge computing in space systems works and share details of suppliers, hardware, and software products on the global marketplace for space.
If you’re familiar with the technology you can use the links below to skip straight to the sections you are most interested in.
Contents
- What is satellite edge computing?
- Commercial space edge computing providers
- Satellite edge computing procurement advice
- Satellite edge computing hardware
- Satellite edge computing software
- Share your requirements
What is satellite edge computing?
Edge computing is the practice of integrating data analysis and processing resources in distributed, physical locations where they can be most beneficial. In space, this has some obvious practical benefits.
You can think of satellite edge computing as a collection of processes, capabilities, and protocols that provide computing resources and data processing capabilities adjacent to data collection payloads – i.e. onboard the satellite.
This approach aims to address the latency, bandwidth, and connectivity challenges that the space environment presents. Broadly, there are two main aspects of space edge computing:
Onboard data processing (OBDP)
Satellites can be equipped with specialized hardware and software that allows them to process data directly on the satellite. This reduces the need to transmit raw data back to Earth.
This can be particularly useful for applications such as Earth Observation, space weather monitoring, and satellite-based internet services.
OBDP approaches can significantly improve the versatility and efficiency of satellites, and even open up new concepts such as hub and spoke constellation models (in which multiple satellites collect data and transmit to a hub satellite, with greater resources and a better orbital position, for processing and downlinking).
Ground-based edge computing
In this approach, satellite data is transmitted to strategically located edge computing facilities closer to end-users or data sources. Such facilities can process and analyze the data before sending it to the central cloud infrastructure or other location, reducing latency and bandwidth requirements.
By combining these two approaches, satellite edge computing can enable more efficient data processing, reduced latency, and improved overall performance for various space applications.
Commercial providers of satellite edge computing products and services
Satellite edge computing products and services are coming to market from suppliers all around the world – here are a range of satsearch Trusted Suppliers offering hardware or software in this area. You can see more information about the individual products and services, from these and other providers, below.
Satellite edge computing procurement
Edge computing in general has grown significantly in recent years, and advances in robotics and artificial intelligence are likely to accelerate this growth further. The global market for space edge computing systems is forecast to be around $839.5 million by 2030.
As the demands for faster and more powerful data processing onboard spacecraft, and reduced latency in data transfer and tasking, have grown, the satellite edge computing supply chain has also evolved around the world.
There are now many computing units with flight heritage, come accruing hundreds of hours of operational time, along with software-only packages from a variety of providers around the world.
Selecting the best option for a mission is therefore a challenge – but here are some key criteria to keep in mind:
Specialization vs. versatility – in general, edge computing systems for space missions that are application-specific will likely provide optimal performance for that application. However, should your mission plan change or if you are designing a mission with multiple objectives, or if you plan to scale up a constellation or service that will require different computing capabilities, you may wish to consider a more flexible option to build on.
Understand the entire data processing chain first – before selecting an edge computing option, ensure you are clear about exactly what form of processing is needed and at what precise step in the data capture, storage, processing, and downlinking chain. This might be a technical judgement, such as needing to remove null data elements prior to compression to maximize storage resources, and/or a more economical one, such as that it may be possible to achieve the results you need with processing on the ground for example, rather than in orbit.
Realistic optimism about scalability – to continue from the previous points, weighing up edge computing requirements shouldn’t just be about what is needed today, but should also factor in what you will require in the future if everything goes to plan. So when looking at the requirements around flexibility and processing chain staging, think about how these would change if throughput was 2X, 5X or 10X the current expectations.
Reliability, security and traceability – understanding how your data is to be handled by different hardware and software systems is vital not only to ensuring consistent operation, but also to secure the system, respond to attacks, and provide auditable proof of the setup. Space is highly regulated and there are many defense and dual-use applications in the supply chain – ensure that you consider how edge computing resources will impact your overall security vectors and feature in both your hardware and software Bill of Materials (BOM) for external review, if needed.
Determine the key value – edge computing is a reasonably broad term in the industry, encompassing several types of solution that can be acquired individually, such as; hardware, software, services, and even algorithms. Ensure you are aware of what specific system you need for your mission.
These are just some of the procurement considerations that go into an edge computing purchase – along with typical factors such as price, heritage, SWAP-C (size, weight, power and cost) budget and so on.
And any good decision should be based on up-to-date, first-hand knowledge of what is available on the market today – which you can find in the next section.
Satellite edge computing hardware products
In this section you can find details of specialized hardware to enable edge computing in space. You can click on any of the links below to find out more about the equipment and the manufacturers on each product page.
From the product pages you can submit free requests for further information, quotes, documents, or whatever else you might need for your procurement or trade study processes.
Alternatively, to quickly send out a free and no-obligation request for quote or proposal to all of these companies, click here and share your requirements with us.
Please note that a wide variety of satellite computing hardware that could potentially perform edge processes, so this list primarily includes systems from suppliers who are specifically focussing on edge applications. If you have any questions or comments, please feel free to contact us.
The Colossus Kestrel is a data processing system designed for CubeSats and small satellites. With the NVIDIA Jetson TX2i module at its core, the Colossus Kestrel adds massive parallel computing capability as well as high-performance CPUs. Colossus' design adds extra storage, including a standard microSD card and M.2 2280 SSD slot. The product's design also provides protection and telemetry circuits for routine monitoring, as well as fault detection and isolation.
Request quote 
request info
The Ibeos EDGE-1100 Payload Processor is a radiation-tolerant system providing on-board processing power ranging from 300 GFLOPS to >1 TFLOPS. A mix of serial interfaces, for communicating with different bus and payload assemblies, are available along with a range of custom configurations.
The GPU002AF is a mezzanine card for high-performance edge computing applications. This product hosts a power-efficient NVIDIA® Jetson™ Orin Nano modu
The Novo Space SBC004AV Single Board Computer has a Xilinx Versal ACAP and can enable AI computing on the edge.
The Ramon.Space NuBox - Space Computing Platform is a modular computing system for space data handling.
NuStream is an ultra-high-density storage recorder for data-driven space applications and space services. Built with in-house radiation-hardened technology, Ramon.Space's storage solution features a high capacity of 1TB, high data rate, and optimized SWaP delivering a reliable and robust solution.
NuPod is a space-resilient, high-performance edge computing platform that delivers smart onboard data processing. This platform comes equipped with Ramon.Space NuStream storage solution and a robust AI DSP machine that enables real-time data analysis, providing actionable information for data-driven missions.
The SkyLabs NANOhpc-obc is designed for most demanding LEO satellite applications with very high-performance needs, featuring fault tolerant 64-bit high-performance multicore RISC-V cluster enhanced with AI accelerators such as Intel® Movidius™ and MicorChip VectorBlox™. This offers a superior AI embedded computer solution for the most challenging applications, that does not take radiation resistance as an option.
The SkyLabs NANOif-2 is a remote terminal unit (RTU) for seamless connection f onboard sensors, instruments or payloads to NANOsky TM/TC architecture. NANOsky TM/TC is an onboard data handling architecture. Several configurable user interfaces, as discrete analog inputs, digital I/Os or more advanced serial communication are supported by NANOif-2.
The Space Inventor Z7000-P3 is an on-board computer containing an FPGA logic system with 125K programmable cells and a dual-core ARM Cortex-A9 MPCoreTM. The chip-based FPGA payload computer provides a wide range of interfaces including LVDS/SpaceWire, and also supports CAN, UART, I2C as traditional OBC interfaces.
The Spiral Blue Space Edge Computer (SE-1) is an onboard computer system designed for Earth observation satellites. It allows Earth observation satellites to process captured images on the satellite itself. The product utilizes NVIDIA Jetson series modules, maximizing processing power while keeping power draw manageable. It also consists of polymer shielding to protect against Single Event Effects and additional software and hardware mitigations.
The Unibap SpaceCloud iX5-106 is designed for space applications. The iX5 family is Unibap’s most power-efficient and reliable computer solution for large and small spacecraft. It combines radiation tolerance and flight heritage, boasting a proven TRL 9 maturity. The iX5-106 model features an AMD Steppe Eagle Quad-core x86-64 CPU and AMD Radeon GPU paired with SATA SSD storage, a Microsemi SmartFusion2 FPGA, and an Intel Movidius Myriad X Vision Processing Unit.
Unibap provides Application Development Systems to enable mission customers to get a head start in their software development, and third-party SpaceCloud® users to create new applications. The iX5 and iX10 Application Development Systems (ODE and ADS-W and ADS-X, respectively) contain the same CPU and GPU architectures as our engineering and flight models with an easy-to-start HW design.
The Amazon Web Services AWS Snowcone is a rugged and highly portable system designed for edge computing and data transfer in extreme environments.
Hewlett Packard Enterprise's compact HPE Edgeline systems are designed to offer rugged enterprise edge computing performance in harsh conditions.
Hewlett Packard Enterprise's HPE ProLiant Servers are designed to optimize data analysis and simplify computing operations in a various applications.
A CubeSat standard-compliant Data Processing Unit (DPU) designed for the application of AI solutions in space. Leopard uses Deep Neural Networks to process data on-board and features FPGA to implement deep learning algorithms. The system has a throughput of up to 3 Tera Operations per second.
The KP Labs' Lion is a data processing unit for advanced operations with the use of artificial intelligence and on-board data processing. The Lion DPU is dedicated to micro and small satellites weighing between 50 and 500kg.
KP Labs’ The Herd - AI-powered algorithms for Earth Observation (EO) - is a set of AI-powered algorithms designed for EO data analysis.
Resilient Computing's RadPC - Radiation Tolerant Computing system is a space computer that uses COTS FPGA tech for in-built failure recovery.
The Skudo Ultimate FPGA Board is a fully functional HSM designed for high-end space and military/defense drone applications.
Spacechips offers a range of re-configurable, multi-band, high-throughput on-board processing transponders and computing systems for edge computing.
The Thales Alenia Space multiMIND is a multi-mission payload processing system suitable for nano and microsat based on FPGA and multi core processing. The system is modular, robust with radhard elements and designed to offer an unprecedented processing power and efficiency. The multiprocessor system also enables further system miniaturization, enhanced system performance and reconfigurability in-orbit.
The Ubotica CogniSAT-XE1 On-Board AI Payload Processor is a PC/104 form factor system offering computer vision & AI compute acceleration.
The Ubotica CogniSAT-XE2 On-Board AI Payload Processor offers Computer Vision and AI acceleration for CubeSats and smallsats.
The Ubotica CogniSAT-CRC Cloud Removal & Compression Solution is an on-board data processing system for cloud detection and removal.
Space edge computing software
In the list below you can software tools, packages, and services for edge computing applications and equipment. You can click on any of the links below to find out more about the system and the supplier.
From the product pages that open you can then submit free requests for quotes, documents, further details, or whatever else you might need for procurement or trade study processes.
Alternatively, to quickly send out a 100% free and no-obligation request for quote or proposal to all of these companies, click here and share your requirements with us.
Insight4EO is Deimos’ on-board processing and autonomy solution for Earth Observation satellites. Insight4EO allows satellites to create and deliver end-to-end EO products autonomously, enabling real-time EO applications. For more specific product generation, analytics applications can be customized (e.g., cloud detection, ship detection, etc).
STORM is SkyServe’s space segment software for onboard data processing across Multispectral, Hyperspectral, and SAR payloads. This enables geospatial analytics from space as a service by deploying AI/ML models onboard Earth observation (EO) missions. STORM comes with a lightweight embedded Linux OS and is a hardware-agnostic platform, which makes it easy to configure across the majority of Data processing units.
The Unibap SpaceCloud® OS is a Linux-based operating system designed for space applications. Together with Unibap's software framework, and a wide application suite, it facilitates simple and reliable execution of Edge Computing, Autonomous Operations, and Cloud Computing in space. SpaceCloud OS’s Linux heritage combined with its reliability and robustness enables rapid software development for a wide variety of users, including those without previous space experience.
The Zaitra SKAISEN is an onboard cloud detection solution designed for Earth Observation missions and a wide range of optical sensors. SKAISEN can save costs associated with data transfer while empowering mission autonomy and supporting its purpose for space missions. Powered by AI algorithms, SKAISEN detects pixels polluted by clouds directly onboard the spacecraft.
The AIKO cloudy_CHARLES is an onboard imagery processing software for cloud segmentation. With CHARLES, the quality of data is assessed directly onboard and only the relevant frames are sent to the ground.
The AIKO orbital_OLIVER is a software solution to enable autonomous spacecraft operations onboard the satellite thereby improving its performance.
The AIKO speedy_SKYLAR is a software tool that enables optimal routing for communications within a satellite constellation. SKYLAR helps in finding a path to send the satellite data back to Earth, even when there is no direct contact with the ground stations.
Edgise's edge AI technologies are designed to bring the power of AI onboard space assets to enhance performance and versatility of the system locally.
IBM Edge is a custom edge computing solution designed to bring autonomous decision-making and data analysis to space applications.
Klepsydra Technologies' Klepsydra AI system is an inference engine designed to run pre-trained AI models on edge devices for real-time applications.
LEOcloud's ground-based edge computing services are designed to provide crucial insights, with low latency, co-located with critical mission data.
Little Place Labs' Space Edge Computing solutions consist of software applications using machine learning techniques for remote sensing satellites.
Palantir's Edge AI software is designed to offer a lightweight and low-power system to run autonomous decision-making on satellites and other assets.
The Red Hat OpenShift® platform is a single solution designed to enables developers to build, package, and run machine learning (ML) models in space.
Nearby Computing's NearbyOne Platform offers end-to-end, cross-domain orchestration capabilities for space edge computing and other applications.
Share your requirements with satsearch
We offer expert procurement support and advice that is 100% free, and no-obligation, for engineers and mission designers across the world.
If you are searching for edge computing capabilities, or are interested in whether the technology can benefit your mission, we can help you refine your needs and contact suppliers across the global market on your behalf.
Our dedicated procurement team is experienced in getting quotes, further information, proposals, documents, or whatever else might be needed to facilitate a new transaction or trade study (for any kind of space-related hardware, software, or service) – typically getting supplier responses in a matter of days rather than weeks.
We can handle multi-vendor conversations and collate all the information in order to share the relevant details with your team. We then hand over the discussion to you to finalize a purchase.
To get started, simply click here and share your requirements with us today. We look forward to hearing from you!
Need help finding the best solution?
Share your requirements with us now
Would you like to be featured on satsearch?
Click here to get started today
Further reading
- Attitude sensors
- Accelerometers
- Gyroscopes
- Inertial Measurement Units (IMUs)
- Sun sensors
- Star trackers
- Satellite components
