NASA's next lunar construction test will not happen on the Moon. It will happen inside a competition arena at Kennedy Space Center, where 50 college teams will try to make small autonomous robots build protective berms out of simulated regolith. The agency's 2026 Lunabotics Challenge runs May 19 to 21, and the assignment is unusually practical: move dirt, shape barriers, collect usable construction data, and prove that robotic systems can prepare a landing zone before astronauts and high-value infrastructure arrive. Student-built construction robots will compete in a simulated lunar surface arena. Credit: AI-generated image A Student Challenge With a Real Artemis Job NASA announced on May 11 that media will be able to attend the 2026 Lunabotics Challenge at the Astronauts Memorial Foundation's Center for Space Education at the Kennedy Space Center Visitor Complex. The competition is scheduled for three full days, from 8 a.m. to 6 p.m. Eastern time, with live streams planned through NASA's Lunabotics page. The event sits inside NASA's Artemis Student Challenges program, but the task is not a classroom abstraction. Each team must design, build, and operate a prototype off-world construction robot. The rover has to work with soil and other material that simulates lunar regolith, then use that material to build a berm, the kind of low protective barrier that future lunar sites may need around landing pads, tanks, power systems, habitats, and work zones. NASA says the competition includes 50 college teams from across the United States. Participants also submit the kind of engineering paperwork that flight projects live on, including a project management plan, a systems engineering paper, a robot data report, a proof-of-life video, presentations, and a working prototype. That mix matters because lunar construction is not just a robotics contest. It is a systems problem involving dust, power, autonomy, scheduling, safety, and repeatability. 50 College teams 3 Competition days 2010 Program start May 19 Final event opens Why it matters NASA is using Lunabotics to test a deceptively hard lunar infrastructure job: getting machines to shape local soil into useful protection before the first permanent surface systems are exposed to rocket plumes, thermal stress, radiation, and dust. Berms Are Boring Until a Lander Fires Its Engine A berm is simple on Earth. It is a mound, wall, or raised barrier made from soil, gravel, rock, or other loose material. On the Moon, that simple object becomes part of the survival layer for a landing zone. When a lander descends or launches, its engine plume can blast abrasive regolith outward at high speed. That ejecta can damage nearby vehicles, sensors, solar arrays, cables, radiators, and pressure vessels. The Moon makes the problem worse. There is no atmosphere to slow particles down. Lunar soil is sharp because it has not been rounded by water or wind. It can cling electrostatically to surfaces, grind into seals, coat optical sensors, and create maintenance problems that Apollo crews saw even during short stays. A long-duration Artemis site will have more hardware, more visits, and less tolerance for preventable contamination. NASA's May 11 advisory lists several possible uses for berms: protecting equipment from debris during lunar landings and launches, shading cryogenic propellant tank farms, helping shield a nuclear power plant from space radiation, and serving other site-preparation roles. That list is a useful snapshot of what the lunar surface is becoming. The problem is no longer only how to land. It is how to keep landing near assets that need to remain operational after the dust settles. AI-generated image Regolith berms could protect tanks, power systems, and nearby equipment from plume debris and harsh lunar conditions. Infrastructure Berm function Risk reduced Landing pads Deflect and contain surface ejecta Plume-driven damage to nearby assets Propellant tanks Provide shade and physical separation Thermal loading and debris strikes Surface power Add shielding mass around fixed systems Radiation exposure and operational disruption Habitats and work zones Create protected operating areas Dust intrusion and equipment abrasion The Engineering Problem Is Small, Dusty, and Unforgiving Lunabotics forces teams to confront the practical edge of in-situ resource utilization. ISRU is often discussed in grand terms: oxygen extraction, lunar water mining, propellant production, and building materials made from local regolith. Berm construction is a more immediate version of the same idea. Use what is already on the Moon, move it with robots, and turn it into protection. That sounds straightforward until the robot has to do it. A useful rover needs traction in loose material, enough power to dig or push, mechanisms that do not jam easily, autonomy that can handle imperfect terrain, and software that can keep working when the surface does not behave like a laboratory floor. A blade that works in one soil bed may bog down in another. A bucket wheel can move material quickly but becomes a liability if it clogs, vibrates, or throws dust into sensitive electronics. The student robots will not experience true lunar gravity, vacuum, or radiation. They will still generate useful data because construction is full of tradeoffs that show up early: how much mass must the robot carry, how fast can it move material, what geometry produces a stable berm, how much operator input is required, and how does the system recover from stalls or poor positioning? AI-generated image A lunar construction rover has to manage traction, dust, power, and autonomy while moving granular material. What NASA can learn from a competition rover • Material handling: How different excavation designs move simulated regolith, and where they fail. • Autonomy limits: How much task planning can be done without constant human steering. • Berm geometry: Which shapes can be built reliably with small mobile systems. • Operations cadence: How long a robot needs to prepare a small protected zone. A Preview of the Lunar Worksite The timing is important. Artemis is shifting from single-mission thinking toward surface systems that must stay useful between crew visits. That changes the definition of infrastructure. A lander, rover, power source, habitat, propellant tank, communications mast, and science payload are not isolated objects. They become a worksite, and worksites need grading, traffic routes, keep-out zones, blast protection, lighting, charging, maintenance access, and safety margins. Robotic construction is attractive because many of those jobs should happen before astronauts arrive. A crewed surface mission has limited time, limited suit consumables, and a long list of science and operational priorities. Asking astronauts to spend major portions of a moonwalk building protective dirt walls by hand would be a poor use of scarce human time. If machines can do early site prep, crews can arrive at a safer and more organized location. There is also a cadence issue. If NASA and commercial partners want repeated landings near the same polar region, every arrival affects the next one. Plume debris can contaminate assets. Dust can degrade solar output. Uncontrolled traffic can make the site harder to manage. Berms, roads, pads, and graded zones are not aesthetic improvements. They are how a landing site becomes a base instead of a campsite. AI-generated image Repeated lunar landings will need prepared zones, not just accurate navigation. Before crew arrival Robots can shape protective barriers and clear zones while astronauts are still on Earth. Between missions Autonomous systems can repair berms, inspect damage, and prepare for the next landing. During expansion Surface construction scales as more tanks, power units, habitats, and work areas arrive. What to Watch at Kennedy The winning robot will not be a flight article. It will not roll onto a lunar lander next month. Stil