A historic computing milestone is set to reach the lunar surface later this year as California-based space company Astrolab prepares to launch its uncrewed moon rover, FLIP, equipped with Hewlett Packard Enterprise’s (HPE) Spaceborne Computer-4. The launch will mark the first time high-performance computing, graphics processing units (GPUs) and artificial intelligence (AI) experiments are deployed and tested directly on the moon. Testing next-gen computational hardware in deep space is viewed as a vital step to support future human habitation and long-term economic activity across the lunar frontier even as Nasa pushes closer to returning astronauts to the moon under its Artemis program.
Edge computing on Moon and what is so special about his computer
The uncrewed rover will touch down near the lunar south pole as part of the Griffin Mission 1 (Griffin-1). Astrolab is one of two private firms selected by NASA to develop a drivable rover for Artemis, as per Space.com.“We’re sending the first supercomputer from HPE to the lunar surface later this year, doing the first GPU on the moon [and the] first AI experiments on the moon. They’re breaking new ground, and they’re ahead of the curve,” said Jaret Matthews, founder and CEO of Astrolab, during a media event at company headquarters earlier this month.To handle the harsh space environment, HPE’s Spaceborne Computer-4 has been engineered as a lightweight, compact, fanless and self-contained system. Its core features include a low-power 64-bit CPU paired with an integrated GPU, solid-state drives (SSDs) by memory specialist KIOXIA, and heat-conductive aluminium housing connected directly to FLIP’s cooling system, with energy supplied by the rover’s combination of onboard solar cells and batteries.However, the primary technical objective of the system is proving that it can survive the grueling environmental extremes of the lunar south pole. Matthews explained: “We have to design for a full year on the lunar surface. So that means surviving multiple lunar nights, upwards of 150 hours of complete darkness where we’re not receiving energy or warmth from the sun.”Matthews noted that the rover must operate autonomously for long stretches when astronauts are absent: “We’re having to develop a semi-autonomous platform that will spend many months [and] travel hundreds of kilometers.”Beyond durability, the supercomputer will actively process visual feeds captured during transit, conduct autonomous observations, assist in terrain analysis and run lunar mapping tasks. HPE noted in a statement that “an AI services container will demonstrate reliable deployment of scalable AI service applications on the moon,” laying the groundwork for how future crewed missions might harness AI in deep space.
Borrowing proven navigation tech from Mars
As the moon lacks a global positioning system (GPS), navigation represents a serious technical hurdle for surface exploration. To solve this, Astrolab is leveraging software originally designed for Mars exploration by NASA’s Jet Propulsion Laboratory (JPL), where Matthews previously worked.“One of the experiments [that will be] running on the spaceborne computer on the moon is a localization software experiment from JPL. It has been used on Mars, and this will be the first time it’s used on the moon,” Matthews added.The system relies on a machine learning algorithm that cross-references onboard map data with real-time visual observations of the surrounding landscape. By matching observed landmarks against its internal memory, the rover can establish its exact coordinates on the map and steer clear of treacherous craters without needing GPS signals.
