EdgeCortix® Inc., a fabless semiconductor leader in energy-efficient AI at the edge, has scored a major milestone in space-ready AI computing. Its SAKURA-II AI co-processor successfully demonstrated high levels of radiation resiliency during NASA’s heavy ion testing, confirming its potential for autonomous applications in low Earth orbit, geosynchronous orbit, and lunar operations.
This marks the second co-processor from EdgeCortix to undergo such testing, following SAKURA-I, which in 2024 set a benchmark for radiation-tolerant AI processing. The NASA Electronic Parts and Packaging Program (NEPP) report confirms that SAKURA-II achieved equivalent performance metrics, with no destructive events and minimal transitory radiation effects during heavy ion exposure at Texas A&M Cyclotron.
Supported by the Defense Innovation Unit (DIU), these tests highlight the growing feasibility of deploying AI-powered autonomy directly in space, from satellite navigation and onboard vision systems to lunar robotics, without the need for power-hungry GPUs or overburdened CPUs.
Why Radiation Resiliency Matters
Radiation presents one of the most unforgiving challenges for electronics in space. High-energy particles can disrupt computations, cause transient errors, or permanently damage processors. For AI systems tasked with real-time decision-making—such as autonomous spacecraft navigation or lunar surface operations—resiliency is non-negotiable.
SAKURA-II’s successful testing demonstrates that energy-efficient AI can operate reliably in extreme environments, opening opportunities for fully autonomous spacecraft and robotic systems that rely on machine learning and computer vision.
Power-Efficient AI in Orbit
Traditional CPUs often struggle to handle the heavy computation of modern AI workloads, while GPUs typically consume prohibitive amounts of power—both limiting the feasibility of onboard autonomy. EdgeCortix’s AI co-processors address this gap with low-power, high-performance AI processing, enabling satellites, landers, and lunar rovers to process sensor data locally and respond in real time.
“The completion of this heavy ion testing and NASA’s published report represent a significant milestone in EdgeCortix’s mission to extend intelligent computing beyond Earth,” said Dr. Sakyasingha Dasgupta, CEO of EdgeCortix. “As space systems increasingly demand autonomy, lower power consumption, and real-time decision-making, SAKURA-II demonstrates that advanced AI processing can be performed reliably directly in orbit and on the lunar surface.”
Implications for Space Autonomy
With SAKURA-II, autonomous space missions can:
- Perform real-time AI-based navigation and decision-making
- Process computer vision and sensor data onboard without relying on Earth-based computation
- Reduce power consumption and extend mission lifetimes
- Increase resiliency in radiation-intensive environments, from satellites in low Earth orbit to lunar exploration
The success of SAKURA-II also reinforces the trend of moving AI computation closer to the edge—even beyond Earth—reducing latency and enabling faster response times for autonomous systems in space.
Looking Ahead
EdgeCortix’s continued progress in space-grade AI co-processors sets the stage for broader adoption of energy-efficient, radiation-resilient AI across the space industry. As missions grow more complex, autonomous capabilities will become increasingly essential, and processors like SAKURA-II could be foundational to the next generation of space exploration.
By combining radiation resiliency with low-power AI processing, EdgeCortix is effectively positioning itself at the intersection of space autonomy, edge AI, and energy-efficient computing—a rare and strategically significant niche.
Power Tomorrow’s Intelligence — Build It with TechEdgeAI
