industry news
Subscribe Now

Aitech Supercharges its Space‑Proven Rugged AI Systems with NVIDIA IGX Thor

Aitech Integrates NVIDIA IGX Thor Platform, Accelerating Rugged AI Edge Computing Solutions and Innovation for Space

Aitech, a global leader in rugged, embedded AI-ready computing solutions, today announced it will integrate the NVIDIA IGX Thor platform into its S-A2300 COTS AI Supercomputer and future designs, expanding AI computing capabilities and ushering in a new era of accelerated computing for space and missions requiring edge AI. By integrating IGX Thor, Aitech is positioned to provide deployable, mission-ready space computing systems for the harshest operational environments.

This milestone reinforces Aitech’s standing as a top-tier rugged computing specialist with more than 40 years of experience delivering high-performance solutions for mission-critical applications, including offering one of the first GPGPUs to fly in space. Aitech now has access to IGX Thor evaluation boards and modules, enabling rapid design, system integration and hardware-in-the-loop testing.

“Aitech is at the forefront of the AI revolution, delivering the most powerful, reliable and future-ready technologies available,” said Dan Mor, director, Customer Solutions, Aitech. “By integrating IGX Thor into our space-proven, rugged AI systems, Aitech is uniquely positioned to help transition advanced AI from terrestrial industrial spaces to the harsh environments of orbit.”

Expanding LEO Capabilities with the S-A2300 AI COTS Supercomputer

Aitech was an early adopter of the NVIDIA Jetson AGX Orin platform, creating space-based AI compute solutions that power missions for NASA, Sidus Space and many other space trailblazers. The S-A2300 COTS AI Supercomputer, which leverages the NVIDIA Jetson AGX Orin Industrial system-on-module, is designed specifically for Low Earth Orbit (LEO) missions. It combines exceptional AI compute density with resilience against radiation and extreme environmental conditions.

Key features of the S-A2300 include:

  • Up to 2,048 CUDA® cores and 64 Tensor cores, delivering as much as 248 TOPS for advanced AI workloads
  • Two scalable, highly configurable NVDLA (NVIDIA Deep Learning Accelerator) engines, simplifying integration and improving portability through standardized deep-learning inference acceleration
  • An architecture optimized for space-based applications, enabling real-time image processing, autonomous navigation, climate monitoring, and space-debris tracking

Already delivering unmatched performance for spaceborne AI processing, the S-A2300 unlocks new levels of operational efficiency and on-orbit intelligence. IGX Thor will expand the capabilities further with up to 5,581 FP4 TFLOPS of AI compute for real-time inference and 400 GbE connectivity through the NVIDIA® ConnectX®-7 SmartNIC. Functional Safety Island (FSI) inside the Thor SoC also enables safety functions through the IGX software stack, ensuring reliable operation in mission-critical environments.

Accelerating Time-to-Mission

With IGX Thor evaluation hardware, Aitech can accelerate development cycles, from concept to validation, dramatically reducing time-to-market for critical space and autonomous missions. Together, IGX Thor and Aitech’s rugged AI platforms are a powerful foundation for the next generation of space-based intelligence and autonomy.

About Aitech

Aitech is the world’s first independent, open-systems COTS/MOTS innovator, offering rugged boards and subsystems that serve as the building blocks for integrated computing and networking solutions. With more than 40 years of experience in aerospace, defense, and space, Aitech delivers mission-proven, customizable solutions across sea, land, air, and space domains. Customers include Airbus, BAE Systems, Boeing, Hindustan Aeronautics Limited (HAL), Israel Aerospace Industries (IAI), Larsen & Toubro Limited (L&T), Leonardo, Lockheed Martin, NASA, Northrop Grumman, Rafael, and Virgin Galactic.

Leave a Reply

featured blogs
May 6, 2026
Hollywood has struck gold with The Lord of the Rings and Dune'”so which sci-fi and fantasy books should filmmakers tackle next?...

featured paper

Want early design analysis without simulation?

Sponsored by Siemens Digital Industries Software

Traditional verification methods are failing today's complex IC designs, which require a proactive, early-stage analysis approach. A shift-left methodology addresses IP block integration challenges and the limitations of traditional simulation and ERC tools. Insight Analyzer detects hard-to-find leakage issues across power domains, enabling early analysis without full simulation. Identify inefficiencies earlier to reduce rework, improve reliability, and enhance power performance.

Click to read more!

featured chalk talk

CBOT MULTIGIG Transceiver Platform
In this episode of Chalk Talk, Anders Thelin from TE Connectivity and Amelia Dalton explore how the CBOT MULTIGIG transceiver platform helps address these challenges with a modular, high-performance approach to rugged, high-density connectivity. We also investigate how this platform supports next-generation system architectures, improves design efficiency, and enables robust performance in even the most demanding environments.
May 18, 2026
8,828 views