NASA Innovator Brian Wilcox on Mars Exploration
by Joe Pappalardo | Submitted Monday Nov 24, 2008 [05:05 AM]
Just as NASA and its Phoenix robot find water on Mars and search for Martian life, a key veteran of several Mars research programs—including 1996's Sojourner (the first Mars rover) and the two rovers currently operating on the Red Planet (Opportunity and Spirit)—is now turning his attention to the moon. Leading NASA roboticist Brian Wilcox was the supervisor of the robotic vehicle group at the agency's Jet Propulsion Laboratory when he proposed a new kind of lunar vehicle, the All-Terrain Hex-Legged Extra-Terrestrial Explorer, or ATHLETE. The system rides on six wheels until it meets an obstacle, then it locks its wheels to step over it. His proposal worked its way through the gauntlet of the NASA vetting process, but Wilcox had a choice: stick with his job as supervisor, or engineer the ultimate moonbot. He chose the hands-on approach, and his efforts are rolling and stomping around in deserts and labs in preparation for a possible trip to the moon. PM caught up with him for a conversation about the past and future of robotic extraterrestrial exploration.
Are you impressed that the Mars rovers Spirit and Opportunity have lasted so long?
The fact that they were designed for 90 days and have lasted for 4.5 years is phenomenal. Many of us who were involved are not too surprised, because once you make something bulletproof for 90 days in a very hostile mission, it's not surprising that they're bulletproof for a very long time.
What did the success of the Mars rover program do for robotic exploration?
A lot of people don't realize that Sojourner was a technology experiment, and at the time it wasn't accepted at all. Even the space community did not particularly believe that it was really feasible to make a vehicle that big—only about 11 kilos—that is essentially a complete spacecraft that has communication and power. The only thing it doesn't have is a rocket engine, but instead, of course, it has wheels and motors. To make something that would survive the very hostile environment of Mars, going through the huge temperature range every day from day to night, people did not really feel that was possible. After Sojourner's success, the whole community had a change in attitude. Suddenly the scientists were saying, “How are you going to get my instrument right next to a rock and put my sensor right on a rock?" This was something that Viking back in the 1970s had been unable to do—so that even though the two Viking landers came down in a sea of rock, neither one could reach a single rock with its 2-meter arm.
Let's talk about the moon. How does NASA call for technologies for a new mission there?
I proposed ATHLETE in 2004 when NASA had a call for proposals for NASA centers, industry and academia. It got 7000 notices of intent, of which it selected 500 invited proposals, and I was fortunate enough to be one of them. I was one of 118 selected in the first round. We got our first round of funding in March of 2005, so by the end of 2005 we built the two complete vehicles that you may have seen in some of the pictures, as well as a third partial vehicle. We used them to do a three-vehicle docking demo in our lab in December of 2005. On the basis of that demo, we were fortunate enough to be one of 38 and were then continued into the next phase. This program was for all technologies, not just mobility or robotics.
What other technology was NASA looking at?
There was resource utilization —turning lunar soil into oxygen—there was advanced structure, advanced electronics, there was advanced propulsion to use methane as a fuel for rocket engines; basically every possible technology that could be applicable to the return of humans to the moon and going on to Mars and beyond. There weren't quotas on specific fields, but I think they wanted a portfolio that had enough work going on in every key area that all the enabling technologies would be covered.
Where does ATHLETE fit in to the return to the moon?
ATHLETE serves really two functions on the moon. First, it gets the payloads off the landers and away from the ejecta of the next lander. The spray of materials comes off right at the impact of landing, so you don't want to be standing anywhere nearby when something lands. The second function is to enable this “Winnebago" mode to allow global-scale exploration, where you have a habitat used as a local base camp that can wander all over the moon and allow you to explore the entire moon. You can have one or two of these habitat shells that can go out and go very long distances—and as long as you have a good solar range, there's really no limit to the distance.
What we wanted was to make something as light as it could be for the payload it carried. Because mass is so important in space life—for every kilogram you launch to the surface of the moon, it takes hundreds of kilograms on earth in a launch vehicle to carry it up there—we wanted to make a very efficient vehicle. So the concept I came up with was to use much smaller wheels than you would normally expect for a vehicle of that size and payload. And if the wheels ever became too small, then it would just lock the wheels with the brakes and walk and use them as feet. So you always have this alternative mode of getting out of a very difficult situation.
What is the range of motion on the legs?
Each leg has six degrees of freedom. Every leg is a complete general purpose manipulator, which is why we refer to them as limbs, not legs. There's also a tool adapter we have on every wheel called a power takeoff; it's a square key that happens to be a ½ in. socket drive, and it rotates with the wheel, so when you use the almost 2-hp motor in the wheel, it also turns that key, and you can clamp tools over that socket drive and actuate any kind of a power tool. Because it's on the end of this limb that has six degrees of freedom, you can maneuver that power tool any way you want. It has the exact same kinematics as the venerable PUMA Robot Arm used in industrial robotics.
How does that compare to the proposed scaled-up version?
The vehicle we want to build for the moon is expected to have a 15-ton payload. The current vehicle has about a 660-pound payload on earth. If you scaled it up, both the weight of the vehicle and the weight of the payload would increase. The vehicle itself weights about 850 kilograms on Earth. The lunar version will have a mass of about 2.5 tons.
Would it work well with other robots?
Admittedly, this is not a highly dexterous robot—we don't expect to do very fine motion with it—but NASA has also developed Robonaut, which is a two-armed, human-scale, five-fingered robot that is very dexterous. So when we need that kind of performance, we can grab that and use the ATHLETE leg as a cherry picker to move that around. When ATHLETE stands up, it's 25 ft., and when it lifts it's leg up, it's another 25 ft., so you can have this robotic astronaut 50 ft. up in the air working on the side of your asset vehicle.
When will you know if ATHLETE will be moon-bound?
The budget profile doesn't call for major investment in lunar surface system assets until about 2014. So the concept is that it would continue as a technology development activity until 2014, unless people become convinced that it's a bad idea. Assuming they don't though, we would continue to develop ATHLETE to a high state of technology readiness. By 2014 we want to do everything that we can reasonably do on Earth to prove that it's a good idea, and that means designing everything as flightlike as we know how to make it. At that point, we will have either convinced people or not, but hopefully convinced people that yes, all the claims of the mass savings and power efficiency are true and everyone believes in the ability to dock and undock payloads. Basically we have to get to a preliminary review stage where a group of very seasoned veterans will say that this is ready to go to flight. So you have to convince people you're using flightlike motors, gears, seals of bearings, all the things that are needed to make a flight system work.
People who read this, also read...
back to top