This Experimental Rover Is Unlike Anything NASA’s Ever Sent To Space

The concept was inspired by the automatons of ancient Greece and the mechanical creatures known as Strandbeests of Dutch artist Theo Jansen.


NASA has never sent a rover to Venus–for good reason. The planet’s surface temperature is so hot, computers, cameras, and sensors can’t function. Since no digital device could survive long enough to gather significant amounts of data on Venus (much less transmit it back to Earth), scientists know little about our neighboring planet.


How do you design an entirely mechanical rover that could survive Venus’s computer-melting conditions? The engineers behind a new Venus rover concept at NASA called the Automation Rover for Extreme Environments, or AREE, looked to an unlikely source: the legendary Dutch artist Theo Jansen.

Theo Jansen advises JPL engineers on the Strandbeests’ logic system that enables them to avoid walking into the ocean. AREE would similarly need a mechanical sensor to avoid obstacles like cliffs and large boulders. [Photo: courtesy NASA/Jet Propulsion Laboratory/Caltech]
Jansen is famous for his Strandbeests, giant mechanical “creatures” that autonomously walk using nothing but the wind and clever mechanical engineering. The artist designed his first Strandbeest in 1990 using just PVC pipe and tape, and he’s continued to dream up new Strandbeests in the decades since. Last year, a major retrospective exhibition on his work opened in Chicago, and is now touring other museums. Aspects of the Strandbeests could be useful to a rover on Venus: They are powered by wind, which is steady on the planet, and they’re also a form of mechanical computer that requires no digital components.

Now, after AREE was chosen to be developed through NASA’s annual Innovative Advanced Concepts competition, the researchers behind it are in the midst of a two-year feasibility study to see if the rover could really work.

The initial concept for AREE featured Strandbeest-like legs before Jansen advised against it. The second concept features a bulldozer-like track. [Image: courtesy NASA/Jet Propulsion Laboratory/Caltech]
Last fall, Jansen came to JPL in Pasadena, California, to help the project’s lead engineer, Jonathan Sauder, and his team develop the rover concept. His first piece of advice was to abandon their initial leg-like design, because the mechanisms he uses are rendered useless if the automaton accidentally flips onto its side. “The beaches of the Netherlands are nice and smooth, but the surface of Venus is rough and harsh,” Sauder says.

Instead, Jansen shared the designs of a caterpillar Strandbeest he was working on. This new beast’s ability to crawl along a sandy beach inspired the team to implement a track system–similar to how a bulldozer has a track around its wheels to help it traverse different types of terrain. The current concept now has a moving track on all four sides of the boxy design so that if it were to tumble down a cliff, it would still be able to keep moving afterward.


The track would enable AREE to recover if it were to go over a cliff. [Image: courtesy NASA/Jet Propulsion Laboratory/Caltech]
 The Strandbeests are automatons–a type of mechanical computer that, as opposed to electronic computers, use a series of gears and pins designed in a manner similar to transistors, with logic gates that open and close based on if certain conditions are met. Purely mechanical computers have a long history dating back 2,300 years to the ancient Greeks, who built a computer called the Antikythera that could predict astronomical phenomenon. During World War II, battleships used mechanical computers to determine where to point their guns, predicting the trajectory of shells and the movement of targets. Clocks are one of the simplest forms of mechanical computers.

Sauder believes that these types of mechanical computers could be key to building a rover that can survive on Venus and collect data without digital sensors or cameras. He says the prototype will be structured around a clock because it provides a slow release of energy, which makes it a good base around which to build more components. One such component could be a device that tracks temperature over time. The clock could also serve as a timer around which to base different actions–like stopping every 30 minutes to collect a geological sample.

Yet the high temperatures on Venus provide an entirely new set of engineering constraints. Right now, the team is trying to build a clock that will work at 450 degrees Celsius–or about 842 degrees Fahrenheit–which is more difficult than it seems. Materials can completely transform under that amount of heat. Aluminum, a common rover material, becomes “Silly Putty,” Sauder says. Springs, a vital component of mechanical computers, have different yield strengths at high temperatures. Materials expand as they get hotter. Even grease–which is used for lubrication for most mechanical systems on Earth–burns or melts. To test their prototypes, Sauder and his team throw them in an oven. He says their first tries “failed miserably.” But once they have completed the clock, they’ll be able to apply the lessons they learned from it to building the rest of the prototype.

AREE uses a radio wave system and Morse code to communicate with an orbiting satellite. [Image: courtesy NASAP/Jet Propulsion Laboratory/Caltech]
As they’ve continued working on AREE’s feasibility, Sauder and his team have realized that mechanical automatons won’t be able to do everything. These types of computers can’t take measurements at a high enough resolution to be scientifically useful. However, some scientists are working on high-temperature electronics made of gallium nitride and silicone carbide circuits that can resist the heat. Current systems feature around 100 transistors on a chip–and, Sauder says, “that’s state-of-the- art.” (For comparison, the iPhone 7’s chip has 3.3 billion transistors.) These high-temperature chips, he explains, are basically stuck in the 1960s because the materials limit their complexity–they can’t be easily miniaturized and they use far more energy. While such a chip could take temperature measurements, it’s far from being able to drive a rover autonomously or send data back to Earth. But Sauder believes that this will improve over time, so he’s designing a rover that will work using today’s capabilities–and could be enhanced, not outdated, by new technology in the future.

The concept is in the very earliest stages of development–the team has a lot of work to do before getting funding and becoming an official NASA mission. Only one concept coming out of the Innovative Advanced Concepts competition has ever made it to mission status. But Sauder feels confident that the technology they’re developing will be useful to any future mission to Venus, even if AREE doesn’t make it that far.


Studying Venus is important, Sauder explains: It’s very similar to Earth in its size and its distance from the sun, and yet the differences between the two planets are striking–and puzzling. Understanding the difference will have implications in the search for life in other solar systems. Both Earth and Venus fit the bill for the type of potentially life-sustaining planets that scientists look for around other stars, yet only one is hospitable to life. Discovering how they came to be so different could help scientists better identify planets that might host life. “Why is Venus a hellacious planet, where Earth is a great friendly place to live?” he asks.

Only a rover that can survive for longer than two hours on the surface of the planet would be able to help scientists answer that question. Sauder hopes his concept will bring more attention to the planet and inspire scientists to study it. After all, Mars isn’t our only neighboring planet with secrets.

The team working on AREE also includes Evan Hilgemann, Michael Johnson, Aaron Parness, Bernie Bienstock, Jeffery Hall, Jessie Kawata, and Kathryn Stack. 

About the author

Katharine Schwab is the deputy editor of Fast Company's technology section. Email her at and follow her on Twitter @kschwabable