It’s three in the afternoon, and in their Nasa lab in Silicon Valley, California, two engineers are playing with a toy designed for toddlers. The melon-sized plaything consists of a tactile lattice of brightly painted beads, connected by wooden rods and elastic cords. It twists and flexes as Vytas Sunspiral and Adrian Agogino crunch it in their hands and throw it between themselves across the room.
One online review describes the gadget as “great for sensory exploration,” but Sunspiral and Agogino are considering it for something way more ambitious: planetary exploration. This bundle of beads, rods and cords, they believe, could form the basis of a new generation of planetary rovers.
“The programme that funded this bit of research, I call it the crazy ideas programme,” exclaims the fast-talking Sunspiral, a towering figure whose name, long blonde hair, beard and glasses suggest a “crazy ideas” lab in Silicon Valley is his natural environment. For obvious reasons, Nasa prefers not to use the word “crazy” in its research but includes the project under its Innovative Advanced Concepts Program. “Here, in the intelligent robotics group,” says Sunspiral, “we do all sorts of advanced research on robots.”
The toy Sunspiral and Agogino are playing with uses what is technically known as a tensegrity system. “It’s a system where, unlike our buildings where everything’s held together rigidly, everything’s held together in tension,” Sunspiral explains. “So you end up with a network of cables that hold rods. You will often have seen artwork that looks like this – kind of crazy weird bars, floating in space.”
“The components are very simple,” adds Agogino, twisting the toy in his hands as he speaks. “They’re just rods and cables connected together, and rods never connect to rods and cables never connect to cables. A child can hit themselves over the head with it and they won’t get hurt. That happens to be great for what we want.”
To prove the point, Vytas encourages me to throw the toy on the floor. Hard. It slams into the floor, squashes together a bit and bounces. “Think of what’s happening when it bounces,” he says. “It’s absorbing energy when it’s going down and it’s releasing energy when it’s going up. It doesn’t break because the energy is distributed.”
You can imagine that dropping one of these on the surface of an alien world would certainly be a lot easier than landing the existing types of robotic rovers with all their precision engineered wheels, motors and instruments. Past missions, such as the Spirit and Opportunity rovers, used airbags to cushion their landings. Because of its hefty size, Nasa’s Curiosity had to be lowered on cables using its elaborate sky-crane system. “Drop one of those rovers 30 feet,” says Agogino, “do you think it’s going to be very happy about it?”
But existing planetary rovers are shaped the way they are for a reason: they need wheels to move around, bodies to hold equipment and arms to deploy instruments or cameras. So how do you do that with a glorified bundle of beads, rods and elastic? “We get a lot of our inspiration from biological systems,” says Sunspiral, who suggests a mechanism that mimics how our own muscles expand and contract could offer one solution. “You could shorten and lengthen the cables to cause the whole thing to roll and move around.”
The instruments would be suspended within the centre of the structure to protect them from damage – cushioned from the impact of landing and locomotion. But they would have to be very different to the sort of boxy equipment carried by existing rovers, capable of being tumbled around and with any arms or probes, safely tucked away. There could even be several of these rovers, tumbling across an alien world and communicating with each other like a swarm of bees.
The rover Sunspiral and Agogino are currently putting their minds to is Super Ball Bot, which they are developing as a concept mission to explore Saturn’s moon Titan. With its petrochemical atmosphere and hydrocarbon lakes, Titan is a fascinating destination and one of the potential candidates in the Solar System for primitive life. The European Space Agency’s Huygens probe parachuted down onto Titan’s surface in 2005 but the cookie-shaped spacecraft had no ability to move around and only managed to return data for around 90 minutes in the hostile environment.
Something like Super Ball Bot would be ideal to survive a descent through the thick atmosphere and cope with movement on the moon’s surface – which includes solid ground, slush and sea. “Titan is one of the most interesting moons,” says Agogino. “We can’t see the surface [from space] and the atmosphere is full of clouds. We really need something robust, or five or six things that are robust, to get the science done.”
So will any of this actually happen or is it just another “crazy” idea?
“Absolutely,” insists Sunspiral. “I’ve been around the robotics industry long enough to know that you just have to throw yourself at a vision and make progress and be successful and anything’s possible. It’s just like Silicon Valley and all the start-ups here. You’ve got to start with a couple of kids in a garage and eventually you have your Google.”
And, I suggest, they are both like those kids in the garage? “Exactly,” says Agogino as they go back to playing with their toy.