Nasa announces new rover mission to Mars in 2020 Options

[Doug M.]

YES!! Really. It's a refly - not a redesign.

I suppose Moore's Law has become so thoroughly internalized that one's immediate, reflexive response to "it's a refly" is "except for the electronics, of course, right? right?" There are of course perfectly good reasons to use heritage electronics! It makes sense. But if you're not a professional in the field, yes, there is a moment of startle.


Doug M.

[mcaplinger]

"Better is the enemy of good enough."

[djellison]

I suppose Moore's Law has become so thoroughly internalized that one's immediate, reflexive response to "it's a refly" is "except for the electronics, of course, right? right?"

When it comes to spaceflight - you would probably change everything else before touching the electronics. Look at Phoenix - it flew with virtual Pathfinder like electronics. The RAD6000 has been flying for 14 years or so. 9 years now since the first RAD750 launch.

[anticitizen2]

Didn't they already fly the spare Marsdial?

[JRehling]

Regarding's Doug's "Really?" to Doug:

Computer hardware has ceased its breakneck rate of accelerating speeds which was a given (Moore's Law) for decades. Commercial products continue to see incremental improvements, but not on the exponential trend which was true until about 10 years ago. For spacecraft, the value of the reliability of a known system outweighs incremental increases in performance.

It would have been unthinkable to use Ranger 7's "computer" on Cassini, but the computers on spacecraft 20 years from now might be about the same as the ones flying now.

Here's one image that sums up the trend. Note the logarithmic vertical axis.

http://smoothspan.files.wordpress.com/2007...clockspeeds.jpg

[craigmcg]

It seems clear though, that for rovers to be more human-like in capability and speed (vs. orbiters and fly-by spacecraft that have much simpler operating modes) advances in robust, reliable computing will be one of the key enablers.

I'm sure it won't be a simple, inexpensive task to design, build and test such capability, and the market for this technology is less than niche.

[vjkane]

When you are trying to fly a new mission for a cost of $1.5B (vs. $2.5B for MSL) you have to find some significant cost savings.

It's a tighter budget than that. Inflation will eat up about a third of that spending amount, so in MSL $s, the 2020 mission is closer to $1B. Tight.

[vjkane]

Commercial products continue to see incremental improvements, but not on the exponential trend which was true until about 10 years ago.

I used to work for Intel. The killer is heat produced by the microprocessors. The engineers could have continued to pump out the performance improvements, but we'd all be using water cooled processors. (Actually have that in my office.) The reason for multicore processors was to go for lower performance per core, but be able to run multiple tasks/threads at once.

[Doug M.]

Regarding's Doug's "Really?" to Doug:

Computer hardware has ceased its breakneck rate of accelerating speeds which was a given (Moore's Law) for decades. Commercial products continue to see incremental improvements, but not on the exponential trend which was true until about 10 years ago.

Yes, but current spacecraft hardware predates the flattening of the curve by a decade or more.

Again, I understand the incentives. It looks like NASA has made a reasonable choice, especially given the cost constraints. But this means that, 20 years from now in the 2030s, there will be NASA engineers poring over the design specs of chips that are older than they are. There's nothing inherently wrong with that picture. Robust technology is robust. (Consider the B-52.) But it takes a slight mental adjustment.


Doug M.

[Astro0]

>engineers poring over the design specs of chips that are older than they are

I can tell you from first hand knowledge that there are already engineers working in this field doing just that.

[nprev]

And in military aviation as well, believe me.

[djellison]

There are people operating spacecraft at JPL launched a decade before they were born. There are people driving Mars rovers that landed when they were in High School.

[Mongo]

Red Dragon would work in conjunction with the Mars 2020 sample collection rover, and be ready two years earlier for far less cost than the previous baseline mission. The above image links to the 70 minute video presentation.

Larry Lemke - Red Dragon: Low Cost Access to the Surface of Mars

Published on Jun 24, 2014

Abstract: One of Ames' long standing science interests has been to robotically drill deeply into Mars' subsurface environment (2 meters, or more) to investigate the habitability of that zone for past or extant life. Large, capable Mars landers would ease the problem of landing and operating deep robotic drills. In 2010, an Ames scientist realized that the crew-carrying version of the SpaceX Dragon capsule would possess all the subsystems necessary to perform a soft landing on Earth, and raised the question of whether it could also soft land on Mars. If it could, it might be a candidate platform for a Discovery or Mars Scout class deep drilling mission, for example.

After approximately 3 years studying the engineering problem we have concluded that a minimally modified Dragon capsule (which we call the "Red Dragon") could successfully perform an all-propulsive Entry, Descent, and Landing (EDL). We present and discuss the analysis that supports this conclusion. At the upper limits of its capability, a Red Dragon could land approximately 2 metric tons of useful payload, or approximately twice the mass that the MSL Skycrane demonstrated with a useful volume 3 or 4 times as great. This combination of features led us to speculate that it might be possible to land enough mass and volume with a Red Dragon to enable a Mars Sample Return mission in which Mars Orbit Rendezvous is avoided, and the return vehicle comes directly back to Earth. This potentially lowers the risk and cost of a sample return mission. We conclude that such an Earth-Direct sample return architecture is feasible if the Earth Return Vehicle is constructed as a small spacecraft. Larry Lemke will present and discuss the analysis that supports this conclusion.

[nprev]

I remember ideas along these lines being tossed around on the Forum a few years back. Glad to see that there's some serious study of the concept in work!

[DEChengst]

Instruments just got announced:

http://www.nasa.gov/press/2014/july/nasa-a...s-never-before/

A bit more detailed information about the instruments is starting to pop up:

PIXL: http://www.jpl.nasa.gov/news/news.php?release=2014-253
SHERLOC: http://www.jpl.nasa.gov/news/news.php?release=2014-254
MOXIE: http://newsoffice.mit.edu/2014/going-red-planet
Mastcam-Z: https://asunews.asu.edu/20140731-mars-2020