Outside of the box

Wall·E, Disney Pixar's portrayal of the prototypical planetary rover. (Credit: Disney/Pixar)

Wall·E, Disney Pixar's depiction of the prototypical planetary rover. (Credit: Disney/Pixar)

Anyone who hasn’t been living under a rock for the past five years of continuous success from NASA’s two Martian surface probes, Spirit and Opportunity, can tell you what a planetary rover looks like: there’s a compartment with electronics, a few solar panels, and maybe some instruments, all carried forward by a set of wheels underneath. This design has served the two Mars Exploration Rovers exceptionally well, but after more than 1750 sols (Martian days) of traverses through the sandy dunes of the Red Planet, the familiar ‘miniaturized car’ concept is beginning to show its weaknesses. On Spirit, for instance, the right wheel ceased functioning on sol 779 after having driven 7 kilometers on the Martian surface and ever since then, its operators back on Earth have had to drive the rover around on the planet in reverse, dragging the dead wheel along.

Perhaps in response to issues such as these, a few weeks ago, a novel concept for a planetary exploration vehicle that quite literally is a thought outside of the proverbial box that the two Mars rovers resemble, surfaced on the web log of open source Google Lunar X PRIZE contestant Team FREDNET: a ball. Instead of a meticulous assembly of little wheels, gears and other delicate and fault-prone moving parts, why not enclose all the electronics safely inside a sealed spherical container and roll it around on the Moon? The concept was originally proposed in the team’s online forums almost a year ago, but had been largely abandoned in favor of more traditional wheeled concepts then making progress. But in late January, a new member came across the idea when going through the archives. His interest was piqued and he decided to have a fresh look at it.

“It is just a rigid ball, designed to move over the surface of the Moon,” explains the new designer, Joshua Tristancho, an aerospace engineer teaching digital electronics engineering at the Universitat Politècnica de Catalunya (Polytechnical University of Catalonia) in Spain. Where Team FREDNET’s other rover concepts are based on conventional wheel drives, Tristancho’s spherical rover moves by continually displacing its center of mass with a motorized weight moving on the interior of the ball, causing the ball to roll. “The ball is half empty, with the rover’s core systems in the other half,” he says. “When the ball is rolling, inertia is controlled by the main drive wheel, allowing the rover to achieve great speeds. The exterior surface of the ball is rough in order to minimize slipping.” He explains that if the rover spins too fast, it will slip like the wheels of a car on a wet surface – only not because of water, but due to the granularity of the fine dust on the lunar surface.

Tristancho was drawn by the simplicity of the ball concept. “The ball is a well-confined thermal system with optimal geometry with respect to incoming solar radiation,” he claims. He explains that a sphere like his rover is subject to the same amount of solar radiation, or insolation, at any given time regardless of its orientation, whereas in more rectangular rovers such as the Mars Exploration Rovers, the insolation depends on the angle of incidence of the sunlight. The uniform insolation of the spherical rover makes it easier to design the vehicle’s internals to cope with the Moon’s highly variable surface temperatures. Furthermore, the ball has no risk of rolling over and getting stuck on one side as the rectangular rovers do.

Once Tristancho rediscovered the concept, it didn’t take him more than a few days to get the ball rolling for his new design, which he dubbed “Picorover”, in reference to its small size. “I was looking for ideas for a rover, and as I was watching my cat play, I began to get an impression of how balance in two dimensions works,” he says. Tristancho’s cat, which reportedly is fat like a ball, throws its weight around in order to keep its balance when playing. “I bought a toy ball and built in the hardware from my old radio-controlled model airplane, and – there it was! The ball concept worked,” he exclaims. “I designed, built, and recorded the first test movie of the Picorover in just three days, mainly because the concept is so simple.”

Since then, Tristancho has been hard at work to refine the concept and scale it to meet the harsh conditions of the lunar surface. He originally envisioned an opaque ball, enclosed in a heat reflective shield, that would be able to stop and open to take pictures of the lunar surface, but concerns over dust ingestion issues as the vehicle opened, as well as competition rules requiring the rover to shoot video while in motion, led him to a yet simpler, fully sealed design in which pictures could be taken through a window in the ball. Due to its sealed design, the Picorover may in principle reach unlimited speeds, unlike wheeled rovers, which must keep their speeds as low as a few meters per minute to avoid accumulation of dust that could otherwise quickly damage or wear down the machinery by abrasion.

To take advantage of this capability, Tristancho intends to integrate a small radar – a so-called nano-SAR – in the rover. Due to the communications delay to mission controllers back on Earth, preventing the team from intervening in moments of high risk, the Picorover will need a clear overview of its vicinity in order to plan operations at the high speeds without driving over ridges or into craters. The software on-board the rover will be designed to exhibit high degrees of autonomy – essentially, mission controllers will supply only a target destination, and the rover will then self-adjust its course as new terrain information becomes available to it during its tumble across the lunar soil. “I hope the nano-radar will give us a range of about 30 meters,” Tristancho says. “We can send directions to the rover, but navigation must be completely autonomous.”

One of the aspects that will be of particular importance in the evaluation of the Picorover is its performance on slopes. Due to its reliance on effectively ‘falling over’ to move, no forces other than the ball’s weight act directly on the ground, and therefore no reactionary force is generated to push the ball forward, as there is in wheeled rovers. This may prove to limit the Picorover’s traction significantly. Tristancho has conducted analyses indicating that apart from the exterior surface’s adherence to the ground, the rover’s performance on slopes depends not only on its weight, but also on how the mass is distributed in the vehicle. In the ideal case in which the center of mass is located on the periphery of the vehicle, the rover would theoretically be able to traverse slopes of up to 30 degrees. “When you go over a slope of more than 30 degrees, your main problem is traction; wheels will slip,” he asserts.

However, Tristancho hopes that the rover’s high maximum speeds can make up for this drawback: “The faster it goes, the steeper slopes could be overcome,” he says. “The most demanding operation is initiating the motion. Once running, the main drive wheel can add some inertia every second, in order to reach a high cruise speed. For this reason, it is important to have a good long-term navigation algorithm, so the inertia can be put to full use,” he explains. “Our two wheeled rovers have a better approach to climbing than the Picorover. They can take their time in order to decide the best strategy of movement. But, on the other hand, the Picorover might make quick decisions which are not as conservative as those of our two other rovers. The ball concept is good for plain regolith, but not for rocky and irregular terrain.”

Regolith is a loose, heterogeneous rock material, blanketing the lunar surface to a depth of several meters, that was formed over the last several billion years’ steady meteoroid bombardment of the lunar surface. Especially the super-fine, highly fragmented top fraction of the regolith, the so-called lunar dust, poses a great challenge to any mission visiting the lunar surface. Apollo astronauts, for instance, reported that once the dust got onto their spacesuits, it clung so tightly to them that it was impossible to wipe off again. One might wonder if a ball rolling carelessly around in this material would not very quickly get coated in a thick layer of dust, much like a cartoon snowball rolling down-hill growing bigger and bigger as it picks up more snow? But Tristancho says he will address this problem by covering the exterior surface of the Picorover in a coating of very thin steel wires that will act similarly to the tread pattern of a car tire. When the rover is rolling at high speeds, the coating will allow any lunar dust that it picks up to be more easily expelled as the rover spins around. Furthermore, the coating will increase the rover’s surface grip as well.

Tristancho has now assembled a small team of student volunteers from his home town of Barcelona, who will help him develop and construct the next prototype as fast as possible. “We are a little late,” he explains. “WRV1 and Jaluro, our team’s two other rover concepts, are in a very advanced state. We have just begun so we must approach this project with fast prototyping techniques to get up to speed,” he adds. “We will develop and test each component separately. Finally, we will assemble the components, do an integration test and then Picorover will be ready for launch; all this has to be done in six months.”

Staying true to its name, Tristancho aims to keep the size of the Picorover equivalent to the size of a so-called CubeSat pico-satellite. CubeSat is a standard for small research payloads weighing no more than 1 kg that is particularly popular in the academic community. Some launch vehicles and larger satellites have payload compartments specifically for CubeSats built-in, allowing the small satellites to ‘piggyback’ into orbit on the launch of a larger payload at relatively low cost. “Our students could compete for one of these opportunities, to prove the Picorover’s reliability in Low Earth Orbit (LEO), in order to qualify the rover’s internal systems, communication links, materials, and so on,” Tristancho says. “Once qualified, these components can also be used in our other rovers,” he adds. Because the surface environment on the Moon is nearly identical to the space environment in LEO in terms of pressure, temperatures and radiation levels, having the team’s systems qualified like this would be a major milestone for the project.

The Picorover will compete against other concepts submitted in the team’s rover design competition, possibly as early as later this year. The best features of all the proposed designs will then be combined into one rover, which will be built and tested, and then hopefully finally sent towards the Moon aboard the team’s lunar transit vehicle. No firm dates are set, but the team aims for a launch before the Google Lunar X PRIZE expires in 2014.

Thanks to ms. Sonia P. Mansilla for her corrections to this article.


About this entry