Slashdot Mirror
NASA's Drone For Other Worlds
An anonymous reader writes: A group of engineers is building a new drone. What sets this apart from the hundreds of other drone development projects going on around the world? Well, these engineers are at the Kennedy Space Center, and the drone will be used to gather samples on other worlds. The drone is specifically designed to be able to fly in low- or no-atmosphere situations. Senior technologist Rob Mueller describes it as a "prospecting robot." He says, "The first step in being able to use resources on Mars or an asteroid is to find out where the resources are. They are most likely in hard-to-access areas where there is permanent shadow. Some of the crater walls are angled 30 degrees or more, and that's far too steep for a traditional rover to navigate and climb." They face major challenges with rotor and gas-jet design, they have to figure out navigation without GPS, and the whole system needs to be largely autonomous — you can't really steer a drone yourself with a latency of several minutes (or more).
A cubesat launcher modified with manoeuvering thrusters so it could do multiple deployments would work. The question is how small and low-power can you make the atomic clocks?
In practice though, You could get away with some lower orbit probes and synchronize your manouevers to only those times they're overhead providing positioning coverage.
On the Moon or Mars they wouldn't reach very far. But a RTG-powered version on Titan would have unlimited range (although may need to land periodically to recharge its flight batteries). And even a rocket or gas jet version would have quite significant range on an asteroid.
Such a design is obviously going to be very mission sensitive, hence the need for different propulsion systems. Some missions would benefit significantly as well from wings to allow for long distance flight on bodies with atmospheres (Venus, Titan, maybe Mars, etc). A couple worlds, such as Titan, might benefit from landing floats. And so forth. But that's where rapid prototyping tech (such as 3d printing) becomes useful - they engineer the base model and then can play around with variants with ease. Hopefully in the end they'll have a sample collector module with a workable version for almost any body in the solar system. And for the interests of science, we really need something like that, a universal adaptable drone module - to be paired with a universal adaptable ion tug module, one of a couple variants of a universal adaptable reentry / landing modules, and the same for adaptable ascent modules.
It's impressive what science can be pulled off on the surface of another world. But it's nothing compared to what we can do here on Earth with a sample return.
Also, I can kill you with my brain.
It seems to me that a rudimentary GPS setup around Mars would be of some use. Why not build that first?
GPS is pretty dependent on ground stations which monitor each satellite in the constellation and provides "corrections" to the whole system for it to work. It also requires a minimum of 3 (usually more like 5) satellites to be in view to come up with a location solution. So, you are talking about a bunch of independent satellites with some remote monitoring capability from the surface to fix positions and keep the system accurate enough. That's a LOT of very expensive infrastructure to get to Mars..
Simpler solution is to just go with a Loran type system. Drop 3 or more fairly simple devices onto the surface that transmit a bit of coordinated RF and you have a solution to easily navigate within the line of site of the transmitters. It's really just the precursor of GPS and gets you pretty good relative location information.
Even better, would be to use a RADAR system and previously obtained map data like many cruse missiles use. It only requires precision terrain elevation data which can be easily obtained from ONE orbiting probe, and allows you to give the "flying" vehicle some autonomy in navigating from point A to point B. As you can obtain more precision terrain information you can further refine your data set and go more places with better and better assurance you know where you are.
"File to fit, pound to insert, paint to match" - Aircraft Maintenance 101
You can do this using exactly the same technology that cruse missiles use, terrain following.
All you need is precision terrain/elevation data and you can create a vehicle that can navigate safely from Point A to Point B using a terrain mapping RADAR. You just match the RADAR image to your known terrain map and voilà, instant location, elevation and orientation information that you use to correct your inertial navigation system. Preplan your probes route and activities and say go to that waypoint, land, do science, and return to base when done.
"But we don't HAVE the necessary elevation data yet!" you say? I thought we where in the process of collecting detailed elevation data from Mars. Plus, if you have even a small amount of data which is detailed enough to land the craft with, you can make flights and collect additional terrain data using the very same RADAR you use to navigate with and expand your detailed dataset further and further away from your starting point over time. So you'd fly using the inertial navigation system past areas you knew, collect data and return to the area you already know, send the collected data to earth for post processing and receive an expanded detailed terrain map in return then do your detailed science in this new area. Wash, rinse and repeat as many times as possible.
"File to fit, pound to insert, paint to match" - Aircraft Maintenance 101