惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

IntelliJ IDEA : IntelliJ IDEA – the Leading IDE for Professional Development in Java and Kotlin | The JetBrains Blog
IntelliJ IDEA : IntelliJ IDEA – the Leading IDE for Professional Development in Java and Kotlin | The JetBrains Blog
C
CXSECURITY Database RSS Feed - CXSecurity.com
博客园_首页
H
Hackread – Cybersecurity News, Data Breaches, AI and More
T
ThreatConnect
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
博客园 - 聂微东
H
Help Net Security
T
Threat Research - Cisco Blogs
Blog — PlanetScale
Blog — PlanetScale
A
Arctic Wolf
G
Google Developers Blog
量子位
U
Unit 42
I
InfoQ
V
V2EX
F
Fox-IT International blog
P
Privacy & Cybersecurity Law Blog
V
Visual Studio Blog
J
Java Code Geeks
大猫的无限游戏
大猫的无限游戏
C
CERT Recently Published Vulnerability Notes
博客园 - 三生石上(FineUI控件)
T
The Exploit Database - CXSecurity.com
T
Tailwind CSS Blog
SecWiki News
SecWiki News
Know Your Adversary
Know Your Adversary
MyScale Blog
MyScale Blog
宝玉的分享
宝玉的分享
The Hacker News
The Hacker News
Project Zero
Project Zero
Application and Cybersecurity Blog
Application and Cybersecurity Blog
月光博客
月光博客
Recent Commits to openclaw:main
Recent Commits to openclaw:main
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
G
GRAHAM CLULEY
C
Cisco Blogs
I
Intezer
Simon Willison's Weblog
Simon Willison's Weblog
O
OpenAI News
Recorded Future
Recorded Future
T
Tenable Blog
W
WeLiveSecurity
腾讯CDC
Stack Overflow Blog
Stack Overflow Blog
T
The Blog of Author Tim Ferriss
www.infosecurity-magazine.com
www.infosecurity-magazine.com
D
Docker
C
Cybersecurity and Infrastructure Security Agency CISA
PCI Perspectives
PCI Perspectives

Universe Today

Just Like Stars, Open Clusters Can Form Binary Pairs Astrophysical Calibration Could "Autotune" Gravitational Wave Detection Something Just Passed Between Us and a Distant Star. When Spacetime Crystallises, a Black Hole is Born The Weirdness of Early Universe SMBHs Gets Even Weirder A Natural Chemistry Laboratory in Protostar Shock Waves A New Model Helps Astronomers Study How Merging Black Holes Ring Why the Second Full Moon of May is a ‘Blue Minimoon’ NASA TESS Reveals Epic All-Sky Map of Distant Worlds Astronomers Observe the Most Chemically Primitive Galaxy in the Early Universe Where Are All the Intermediate Mass Black Holes? Microlensing Fast Radio Bursts Might Reveal Them When the Sun Tries to Explode and Fails The Sun Just Did Something Nobody Expected and it Kept Going For 19 Days Three Stars, One Extraordinary System and a Drama Still to Come The Definitive Census of Multiple Star Systems Within 10 Parsecs Are Satellite Megaconstellations Accidentally Geoengineering the Earth? The Risk of Stellar Flybys and GJ 710 How Mars Can Help Us Understand 'Marginal' Exoplanets Ultrahigh-energy Cosmic Rays May Be Ultraheavy in Origin
NASA's Next-Generation AI Processor Passes Early Testing
Matthew Will · 2026-05-26 · via Universe Today

As space exploration takes spacecraft and crews farther and farther from Earth, the need for systems that can operate with minimal (or zero) human oversight increases. This is needed in part because of the distances involved, where astronauts on the Moon and Mars will be subject to communication delays that make it harder to send and receive reports and data promptly. At the same time, autonomous systems have the potential to accelerate the rate of scientific returns through faster data analysis.

At the same time, processors need to be hardened to withstand the harsh environment of space, including extreme temperatures and high levels of radiation. To address this need, NASA's Game Changing Development (GCD) program entered into a commercial partnership to develop the High Performance Spaceflight Computing (HPSC) system. This revolutionary chip will give spacecraft the processing capabilities to think for themselves, and just completed its first round of environmental tests.

Building computers that can handle the challenges of spaceflight requires processors that deliver high-quality data while also withstanding the harsh conditions of space (extreme temperature variations and radiation). Traditionally, NASA has relied on chips developed in previous decades, which were designed to be hardy and reliable. But given the current rate at which missions accumulate data, upgraded chips are needed to process and analyze it faster and more efficiently.

NASA's HPSC shown with its exterior casing attached. Credit: NASA/JPL-Caltech/Ryan Lannom *NASA's HPSC shown with its exterior casing attached. Credit: NASA/JPL-Caltech/Ryan Lannom*

This has led NASA and the Arizona-based tech company Microchip Technology Inc. to create HPSC, a radiation-hardened high-performance processor capable of delivering 100 times the computing capability of current systems. The project is managed by the Space Technology Mission Directorate (STMD) based at NASA Langley. NASA's Jet Propulsion Laboratory (JPL) developed mission requirements, funded industry studies, and selected Microchip Technology as its commercial partner, which funded its own research and development.

The HPSC processor is an example of a system-on-a-chip (SoC), a compact integrated circuit typically found in smartphones and tablets that integrates all essential elements of a computing system into a single microchip. This includes central processing units (CPUs, computational offloads, advanced networking units, memory, and input/output interfaces.

However, only the SoCs JPL is testing are built to survive for years while operating millions or even billions of km from Earth and repair facilities. “Building on the legacy of previous space processors, this new multicore system is fault-tolerant, flexible, and extremely high-performing,” said Eugene Schwanbeck, program element manager in NASA’s GCD program. “NASA’s commitment to advancing spaceflight computing is a triumph of technical achievement and collaboration.”

In addition to computing power, the chips will provide high-performance AI dataflow processing with scalable vector computing capabilities. In addition, the HPSC is designed to be adaptable in terms of power consumption and computing performance, so functions can be turned off when not in use or put into lower-power modes. This ability means the HPSC processor can be used for missions with varying power requirements, improving power efficiency since electrical power is such a vital resource in space.

Animation of a lunar rover operating autonomously using NASA's HPSC chip. Credit: NASA/JPL-Caltech *Animation of a lunar rover operating autonomously using NASA's HPSC chip. Credit: NASA/JPL-Caltech*

To ensure these chips can handle spaceflight, technicians at NASA’s Jet Propulsion Laboratory have been conducting various tests that replicate the challenges they will face, such as electromagnetic radiation and extreme temperatures. High-energy particles from the Sun (solar wind) and cosmic rays can cause errors, and extreme cold can cause batteries to shut down, both of which will send a spacecraft into “safe mode." Said Jim Butler, High Performance Space Computing project manager at JPL:

We are putting these new chips through the wringer by carrying out radiation, thermal, and shock tests while also evaluating their performance through a rigorous functional test campaign. To simulate real-world performance, we are using high-fidelity landing scenarios from real NASA missions that would typically require power-intensive hardware to process huge volumes of landing-sensor data. This is an exciting time for us to be working on hardware that will enable NASA’s next giant leaps.

Testing began in February at JPL and will continue for several months, and the results have been promising so far. From what JPL engineers have seen, the processor is performing at 500 times the performance of the radiation-hardened chips currently in use. Once certified for spaceflight, NASA will incorporate the chip into the computing hardware of future orbiters, rovers, habitats, and deep space missions.

In addition to space, the HPSC chip has applications in Earth-based industries such as aviation and automotive manufacturing, for which Microchip Technology is currently adapting it. The HPSC's versatility is in keeping with NASA's history of developing cutting-edge technology while also providing "spinoffs" benefits for numerous fields on Earth.

Further Reading: NASA