NASA'S ERNEST ROVER COVERS 16 MILES IN DESERT DESERT, 10X FASTER

A prototype rover just did something that would be a milestone for planetary exploration. It covered 16 miles of rough terrain in 37 hours, more than ten times the speed at which NASA's Perseverance rover can navigate on Mars.

The rover is called ERNEST, short for Exploration Rover for Navigating Extreme Sloped Terrain. It is a prototype developed at NASA's Jet Propulsion Laboratory in Pasadena, California, which is managed by Caltech. In March 2026, a team took the four-wheel machine out to the Colorado Desert near Plaster City, California, and ran it through its paces. The goal was not just to see how far it could go, but to test autonomy software designed for a future lunar mission that would require far higher speeds and greater mileage than anything JPL has flown before.

you may like

NASA PICKS CREW FOR ARTEMIS III

NASA'S ASTROPIX GAMMA-RAY SENSOR TO FLY ON 2027 ROBOT ARM MISSION

TSMC ADDS $100 BILLION TO ARIZONA CHIP BET, TOTAL HITS $265 BILLION

THE TEST THAT MATTERS

The 16-mile traverse is the headline statistic, but what it represents is more important. Current Mars rovers like Perseverance drive at a cautious crawl, stopping frequently to assess the terrain with onboard cameras and wait for instructions from Earth. That works for a stationary science platform on a planet where a round-trip radio delay is several minutes. For the Moon, where the delay is only a couple of seconds but the terrain is steep, shadowed, and unforgiving, a rover that never breaks out of a gentle jog is not going to cover the distances needed for high-priority science.

ERNEST's test drove continuously over rocky desert terrain, relying on its autonomy software to pick paths, avoid obstacles, and keep moving without a human in the loop. The team followed in chase vehicles, monitoring the rover as it covered ground more efficiently than any previous JPL prototype. The 16 miles, accumulated over 37 hours, is a benchmark that directly scales to what a lunar mission might demand.

WHY THE SPEED GAP MATTERS

ERNEST averaged about 0.43 miles per hour across its 37-hour run, and that number is misleading because it includes stops for testing and observation. The peak sustained speed was higher. The point is that the autonomy stack works at a velocity that changes the mission calculus. A rover that can cover several kilometres per day, rather than a few hundred meters, opens up targets that were previously out of reach: crater rims, lava tubes, polar deposits, the types of features that the next generation of lunar science objectives targets.

DAY, NIGHT AND DAWN

The team did not stop at daylight driving. They ran ERNEST at dusk, dawn, and through the night, simulating the long, low-angle shadows found in the Moon's polar regions. The lunar poles have permanently shadowed craters where sunlight never reaches, and any rover exploring there will spend a lot of time operating in near-total darkness. ERNEST's test at night tested how its sensors and autonomy software handled those conditions, with team members setting up illuminators on the rover to replicate the way a real mission might use headlights or floodlights. The results are not broken out in the public data, but the fact that the test included night runs at all tells you that JPL is serious about a polar mission.

FROM INTERNAL EXPERIMENT TO PROGRAM

ERNEST started as a skunkworks project in 2022, funded by JPL's internal research and development dollars. That is how a lot of NASA's best prototypes are born: an engineer with an idea, a small pot of money, and a mandate to prove the concept before the agency commits to a formal program. By 2026, the project had attracted funding from two separate NASA offices: the Mars Exploration Program and the Exploration Science Strategy Integration Office under the Science Mission Directorate in Washington. That shift from internal experiment to program-funded is a strong signal that the agency sees ERNEST's autonomy capabilities as relevant to both Mars and the Moon.

The dual funding is telling. The Mars program is interested because faster traverse speeds would let a future rover reach multiple science targets in a single mission, increasing the return on investment. The Exploration Science office is interested because the same autonomy could support the human-exploration infrastructure planned for the lunar south pole, where crews will need robotic scouts to survey landing zones and resources before astronauts arrive.

WHAT THIS ACTUALLY PROVES

A single field test does not prove that the autonomy software is ready for a planetary mission. It proves that the software works on Earth, on terrain that is an analogue for the Moon but not an exact match. The next step is to prove that the system can handle the thermal, radiation, and vacuum environments of space, and that it can survive the landing loads that would precede any traverse. Those are hard problems, but they are engineering problems. The core autonomy question, can the rover drive itself over unknown ground at useful speeds without getting stuck or crashing, appears to have been answered.

The team observed the rover with a wheel up on a rock during one of the test runs, a reminder that even advanced autonomy occasionally needs to drive over obstacles rather than around them. That is fine. The test's success is measured in miles, not perfection.

FASTER DATA FROM SPACE

The funding from both the Mars and lunar exploration offices suggests that the technology is being positioned as a candidate for whatever comes next: a Mars sample return campaign that needs a faster fetch rover, or a lunar south pole science mission that needs to cover a hundred kilometres across multiple days.

what to read next

HOT WHEELS INFINITE RUSH PULLS AHEAD TWO WEEKS, LANDS SEPTEMBER 10

BROADCOM SHARES JUMP NEARLY 4%, APPLY CHIP DEAL THROUGH 2031

META PAYWALLS CONVERSATION FOCUS ON SMARTGLASSES FOR $20/MONTH

For now, the 16-mile desert run is the data point that matters. Planetary rovers have crept for fifty years. ERNEST showed that they can finally move.