Lunar soil acts as insulation, reducing energy to keep fuel cold
At the edge of what humanity has yet to build, a group of university students in Virginia has designed a machine meant to do something civilization has never accomplished at scale: reshape the surface of another world for human habitation. Their robot, entered into NASA's Lunabotics Challenge, excavates the moon's fine, ancient soil and piles it into berms — barriers that would shield launch pads, insulate fuel, and guard astronauts from radiation. It is a small machine carrying a large idea: that permanence beyond Earth begins not with grand gestures, but with the patient, unglamorous work of moving dirt.
- Lunar soil behaves unlike anything on Earth — fine as baby powder, abrasive as glass, and unpredictable under the mechanical stress of excavation, making every design choice a gamble.
- The UVA team had no dedicated testing facility, so they ran their robot through a beach volleyball court, using sand as a stand-in for a surface 384,000 kilometers away.
- Their machine must not only dig but navigate, transport, and deposit material with precision — all in service of building the infrastructure that would make a moonbase survivable.
- In sand trials, the robot reportedly outperformed last year's Lunabotics winning score by more than double, a result that suggests a genuine engineering breakthrough rather than incremental progress.
- The competition at Kennedy Space Center now stands as the immediate proving ground, but the deeper stakes are the Artemis missions themselves — and the question of whether humans can truly stay on the moon.
At the University of Virginia, a student team has built a robot for NASA's Lunabotics Challenge — a competition held at Kennedy Space Center that tasks teams with solving one of the central problems of sustained lunar exploration: excavating the moon's surface and using that material to construct protective berms.
The reasoning is practical and far-reaching. Berms built from lunar regolith would reduce the dust thrown up during spacecraft landings, protect stored fuel from the harsh lunar environment, and — perhaps most critically — provide radiation shielding for astronauts living on the moon for extended periods. Cryogenic propellants buried under regolith stay cold more efficiently; habitats covered in lunar soil become safer places to live. The robot, as team lead Craig Kalkwarf frames it, is not merely a digger — it is a tool for making another world habitable.
The engineering demands are formidable. Lunar soil is nothing like terrestrial sand. Its particles are extraordinarily fine and light when undisturbed, yet compact densely under pressure, and its abrasive quality degrades machinery in ways Earth materials do not. The team tested their machine in the university's beach volleyball court while a dedicated regolith facility — funded by an $86,000 grant — remains under construction.
The results from those sand trials have been striking. Kalkwarf estimates the robot would score more than double the highest mark from last year's competition. Whether that performance holds at Kennedy Space Center remains to be seen, but the deeper implication is already visible: somewhere between a volleyball court in Virginia and the lunar surface, students are quietly laying the groundwork for the infrastructure of human permanence beyond Earth.
At the University of Virginia, a team of students has engineered a robot designed to do work that no machine has ever done at scale: mine the moon's surface and use that excavated soil to build protective barriers for future human habitats. The machine is their entry into NASA's Lunabotics Challenge, a competition held at Kennedy Space Center that pushes teams to solve one of the central engineering problems of sustained lunar exploration.
The challenge this year is straightforward in concept but demanding in execution: extract lunar regolith—the fine, powdery soil that blankets the moon—and use it to construct a berm, a raised earthwork that can serve multiple purposes. Craig Kalkwarf, the team's senior mechanical lead, explains the practical reasoning. Those berms would shield the area around spacecraft landing pads, reducing the amount of dust thrown up during descent. They would protect stored fuel and other critical infrastructure from the harsh lunar environment. And they would provide something even more fundamental: radiation shielding for astronauts living on the moon for extended periods.
The engineering challenges are substantial. Lunar soil is not like sand on Earth. It has a consistency closer to baby powder—extremely fine particles, light and fluffy when undisturbed, but capable of compacting densely under pressure. It is abrasive in ways that Earth materials are not. It behaves unpredictably under the conditions that a digging machine must operate within. The robot must excavate this material efficiently, transport it, and deposit it in a controlled manner, all while navigating terrain that astronauts themselves will find treacherous.
Beyond the immediate function of dust control, the berms represent something larger: the infrastructure of permanence on another world. Cryogenic propellants—the fuels that power spacecraft—must be kept cold. If those tanks are buried under regolith, the lunar soil acts as insulation, reducing the energy required to maintain their temperature. Human habitats, similarly, could be covered with regolith to shield occupants from the constant bombardment of radiation that sweeps across the lunar surface. The robot, in other words, is not just a digger. It is a tool for making the moon habitable.
The UVA team tested their machine in an unconventional location: the university's beach volleyball court. A dedicated regolith testing facility, funded through an $86,000 grant from the Jefferson Trust, is still under construction. Working with sand as a proxy for lunar soil, the students put their robot through its paces. The machine drives across terrain, excavates material, and dumps it in the designated area. According to Kalkwarf, the performance has been striking. If the robot performs at the competition as it did during sand trials, it would score more than double the highest score from the previous year's competition.
That projection carries weight because it suggests the team has solved, or at least substantially advanced, the core engineering problem. They have built a machine that can handle the peculiar demands of lunar excavation. They have designed it to operate efficiently under conditions that will only become more familiar as NASA's Artemis missions push toward sustained lunar presence. The competition itself is a proving ground, but the real test will come when machines like this one—or descendants of it—actually touch the moon's surface and begin the work of building the infrastructure that will allow humans to stay.
Notable Quotes
If we store cryogenic propulsion propellants on the moon, we can cover them in regolith, which insulates them, keeping them cold, requiring less energy.— Craig Kalkwarf, senior mechanical lead
Lunar soil is like baby powder—very small particles, very light and fluffy, but then it gets dense fast.— Craig Kalkwarf
The Hearth Conversation Another angle on the story
Why does the moon need berms? Why not just land somewhere else?
Because we're not just visiting anymore. Artemis is about staying. You need fuel depots, habitats, equipment. A landing pad kicks up dust that damages everything nearby. A berm stops that. It's basic infrastructure.
And the radiation shielding—how much regolith would you actually need?
Enough to bury a habitat. The moon gets hit constantly by cosmic rays and solar radiation. A few meters of soil overhead cuts that exposure dramatically. It's the same principle as a fallout shelter, but permanent.
The robot tested in a volleyball court. Does sand really behave like lunar soil?
Close enough to learn from, not close enough to be certain. That's why the competition matters. It's the first real test in something closer to actual regolith simulant. The team's projections are bold, but they're based on actual performance data.
What happens if the robot fails at the competition?
They learn what breaks and why. But the bigger picture doesn't change—someone has to solve this problem. If not UVA this year, then another team, or NASA itself. The moon won't build itself.
Is this just a competition, or is this actually going to the moon?
It's a competition that matters because it's solving a real problem NASA needs solved. The designs, the lessons, the innovations—those feed into actual lunar missions. It's not theoretical. It's engineering that will be tested on Artemis.