Current missions to Mars have been complex affairs, typically using parachutes to guide robots and their delicate scientific instruments down to the surface. In 2012, things got even more high-tech when the space agency used a hovering platform to lower the Mars Rover onto the planet.
In contrast, prototypes of the new robot resemble something a child might make in a craft class -- one was a mass of wires, control cables and Lego bricks -- but what it lacks in appearance it more than makes up for in physics.
(See video of the robot in action on IDG TV.)
The six rods and 24 cables that link them are under a mix of tension and compression make the device incredibly resilient to impacts.
"We've decided that these structures are so strong that they can actually withstand the landing at 30 miles an hour without any additional support," said Adrian Agogino, a scientist working at NASA Ames research center. "So no additional airbags, no additional landing support. This entire thing could actually land on Titan and then, as an active structure, could roll around like a rover."
Researchers have thought about using such so-called tensegrity structures for some time, but have lacked complex computer algorithms that can model how they work, but that's now available and the results are encouraging, said Agogino.
Being robust enough to survive a drop onto a planet also means the robots are plenty robust enough to roll around a planet's surface without getting into trouble.
They move by adjusting the tension in the cables, which has the effect of changing the shape of the robot so its balance shifts and it falls to one side.
Existing robots are very carefully controlled because one small mistake could literally halt a multi-million dollar research program.
"If you have a very robust robot, you don't have to care about that so much," said Agogino. "If it's robust enough that it just won't get into trouble, you just send it going. Maybe it hits a rock, maybe it hits some uneven terrain, you don't care. It's already survived a high-altitude landing, it has enough degrees of freedom to get itself out of trouble."
He envisages the complex scientific instruments being held at the center of the device and probes extending to the edge.
Should it all work, Agogino has an intriguing idea for what it might make possible,
"My fantasy, I don't know if it ever will be done, is you actually send two of these. One of them you have controlled by NASA operations, the other one you just have it controlled on the Internet. You have people tell it where to do and since it's structurally robust, people can make mistakes and give it bad commands but it should survive that. I'd like to see the return on the Internet scientists versus the NASA scientists."
There is still a lot of research that remains to be done in this project. NASA estimates it's at least a decade away from making it onto a mission.
Martyn Williams covers mobile telecoms, Silicon Valley and general technology breaking news for The IDG News Service. Follow Martyn on Twitter at @martyn_williams. Martyn's e-mail address is martyn_williams@idg.com