http://www.nasa.gov/news/media/newsaudio/i...ml#.UdsrNHZ385a

Tuesday, July 9 at 3 p.m. EDT: NASA Teleconference on Mars 2020 Plans

NASA is hosting a media teleconference to provide details about a report that will help define science objectives for the agency's next Mars rover.

The report, prepared by the Mars 2020 Science Definition Team (SDT) NASA appointed in January, is an early, crucial step in developing the mission and the rover's prime science objectives.

Anyone have a link to the report? It's supposed to be at http://mars.jpl.nasa.gov/m2020/ but I can't get past all the fun and games!

http://mepag.jpl.nasa.gov/reports/MEP/Mars...eport_Final.pdf

and images for the briefing have appeared http://www.nasa.gov/mars/telecon20130709/#.UdxZ3uCmW01

Ouch! Keith Cowing poses the tough questions as usual....

I am so glad that they are putting a cache on this rover, although the chances of NASA funding a sample return mission are extremely low. Unless, of course, this rover actually finds a rock unit with high suggestive evidence of past or present life.

The nice thing about a cache is that it has no time limit on it. The mission to come get the cache could take place 2, 4, 20, or 80 years later. So the funding pathways are pretty forgiving in that respect.

Absolutely right! It doesn't have to be any part of the plan now.

Phil

According to an article in Aviation Week,

“While the two rovers will have many components in common, including the chassis and a radioisotope thermoelectric generator for power…”

Can anyone confirm that MSL2 will have a plutonium RTG, like Curiosity ? (Thnx)

http://www.aviationweek.com/Article.aspx?i...3-01-595935.xml

It's assumed to be an RTG at this point, but a final decision has not been made. If you notice - the blueprint-style artwork show neither an RTG nor solar arrays.

Read the SDT report and the FAQ at http://mars.jpl.nasa.gov/m2020/

"No final decision on a power source for the 2020 rover would be made until the mission completes a review
through the National Environmental Policy Act process, which considers the environmental impacts of launching
and conducting the mission. This process is currently scheduled to conclude in late 2014. The baseline-design
power source for 2020 mission planning is the same as Curiosity's: a multi-mission radioisotope thermoelectric
generator. Other possible power sources are also under consideration, including solar power."

In my former life as an environmental consultant I used to write NEPA documentation (mostly for local transportation projects). To comply with NEPA, NASA has to write an Environmental Impact Statement that (among other things) considers nuclear and solar power options for the mission and demonstrates that the impact of using solar power for the mission would be sufficiently harmful to the mission goals as to overcome the slightly greater launch disaster risk posed by the use of an RTG. Which means that, at least in public, NASA cannot officially state which type of power supply they plan to use until the Environmental Impact Statement has been drafted, commented on by the public, those comments replied to, and the statement formally approved.

How can ANYONE possibly justify going through a REPETITION of this lengthy & costly process for MSL2, after it was completed for MSL1, absolutely boggles the mind !
No wonder NASA never has enough funding !

How lengthy or costly do you think it was? I believe that most of the work to qualify the MMRTG was done long before MSL. You can get more insight into this by reading http://science.nasa.gov/media/medialibrary...L-FEIS_Vol1.pdf

At any rate, those are the rules.

Precisely. There's no choice here, and there's no point in bemoaning it. Moving on.

sorry to resurrect this old thread: wasn't the call for instruments to be put out by late fall this year?
I have heard nothing on the subject.

It happened I'd say, see here.

A few people I know prepare alot of paperwork lately. COIs should enlist and give their role in the new instruments.
So the instrument proposal train is still on track.

thanks! I really missed that one!

Instrument proposals for 2020 were due a half hour ago. Now to see what gets selected. We put in some interesting proposals, about which I can say nothing more.

The selections are supposed to be announced in April.

I will be curious to see if this rover is able to get from point a to point b more rapidly. I'm talking about the net speed including all of the following:

- planning cycle (driven by a number of factors, including communications delays and the differences between mars and earth time)
- power available for driving
- maximum wheel speed
- effectiveness of auto navigation and our trust in it
- willingness to drive by things that might be interesting

Obviously terrain will also has an impact, but that is a factor external to the capabilities of our human / machine roving systems.

The Science Definition Team's (SDT) report spent a fair bit of time on the question of how to balance time between traveling to new locations and identifying and collecting samples to cache. It takes a considerable time to assess and area and to plan and then acquire samples.

Here are some sample tradeoffs from the SDT's presentation to MEPAG.

Maximize quantity of cached samples
5 km total driving
34 samples
2 cores per characterized target

Maximize non-caching science
3 km total driving
20 samples
full complement of fieldwork (1 core per characterized target)

Maximize driving
15 km total driving
20 samples
2 cores per characterized target

If you expect any of those to change for Mars 2020, you are going to be disappointed.

The ideal situation would be to use HiRISE images for route planning, with a rover robust enough that it could cope with sub-HiRISE obstacles and use visual odometry to ensure it was remaining on track. This is for a post-2020 rover, methinks, but we will never see 1000 m/day traverses without something like this.

Phil

I guess to save roughly $1 billion in mission costs, you have to make some trade-offs!

It will be interesting to see how fast rovers can improve speed of operations (in general, not just speed of travel). It would also be interesting to see the economics of making a smarter, faster, more expensive rover (and supporting technology infrastructure) vs. the cost of all the human operations.

My recallection is that the cost of bulilding and launching rovers is several times the yearly operating costs. So operating cost is already low.

“If you expect any of those to change for Mars 2020, you are going to be disappointed.”

What makes you say that, Doug? Kindly explain. The SDT and its appendix already make a lot of recommendations which –to me at least- could pave the way towards a next level in efficiency in operations. On top of that, the ongoing, complex MSL operations looks to me the perfect “incubator” for maturing new operations concepts and –strategies out of the hard-earned experience, so slamming the door on positively formulated 2020 hopes/aspirations seems odd If I may say so.

They haven't even settled on a power source: solar or RTG? Until then we can't really make driving comparisons between a rover that exists on Mars right now and one that won't for many more years, anymore than we can compare their scientific instruments!

Most of the suggestions Doug was reacting to were for changes to hardware, not operations.

I don't know how you're going to change the planning cycle very much, as it's dependent on the relative diurnal cycles of Earth and Mars and how much time it takes to do planning.

Also, the mission is highly cost-constrained and any change that costs money is probably DOA.

The real tradeoff is between time spent doing science/caching and driving. The goal of the rover is to do science and caching with driving serving the purpose of getting to the next interesting science/caching location. If operations improve driving distance per day by, say, 2X, they'll get get to the science/caching stations faster and spend more time at them.

- planning cycle
There is nothing that's going to change that. It'll still take a full working shift to plan a rovers day of activities. It's going to basically be the same vehicle, so it's going to be basically as complex to use.


- power available for driving
Will almost certainly be a similar MMRTG to MSL. Anything else would require significant (and expensive) changes to the vehicle design. The waste heat from the MMRTG on MSL is an integral part of the entire thermal architecture of the vehicle. Anything else would be a huge alteration that would throw any notion of build-to-print out the window.

- maximum wheel speed
You'll be seeing the same suspension, the same motors. It'll be doing the same speed.

- effectiveness of auto navigation and our trust in it
2020 will almost certainly fly the same avionics with the same performance as MSL. There may be mild improvements in Autonav - but the code is about as good as it's going to get - it's been running on Mars for years and years (more than half a decade on MER before MSL)

I don't think there's any question that the actual rover will be an RTG. But under federal environmental law, they cannot go forward developing the mission as nuclear-only until they have produced their Environmental Impact Statement (which must be prepared as a draft first, put through a public comment period, and then published in final form) showing that a solar-powered rover would be dramatically inferior to a nuclear-powered one, and therefore the (negligible) risks of building and launching an RTG are justifiable. If you're curious, you can check out the EIS for MSL at this page. The three alternatives (nuclear powered, solar powered, and the "no build" option) are described in chapter 2.

I assume since Mars 2020 will use the same EDL landing system, the landing ellipse size will be more or less the same size as the one on MSL. There might be some room in the system to shrink it slightly. In my opinion there should be some improvement in balancing the landing ellipse location and any primary mission destination's. Especially if the majority or the whole mission is spent driving to it. Was the MSL landing ellipse the absolute best the team could do based on safety and how close you could get it to the destination they wanted to go?

At some point in the future, I assume a more capable rover will be able to be given a rough set of waypoints (for a day's travel) and do most of the navigation by itself, taking a few pictures / samples / measurements along the way and reporting on them when a communications link was available. It would have a "rolling horizon plan" so that it would never have to wait for home base to tell it what to do.

I am curious what a roadmap to this kind of capability would look like in terms of steps and years of development.

Yes. Indeed if you look at the ellipse from before launch, and the ellipse just before landing you'll notice that it both shrunk AND moved closer to the target based on in-flight performance of the vehicle. They didn't leave anything on the table - they did the best the system could reliable support.

How far they have to drive is obviously affected by whether there's science available inside the ellipse or not. I remember from the last landing site selection meeting that there were a lot of scientists who were very concerned about the so-called "go-to" sites (of which Gale is one), because they were worried about problems leading them to not getting out of the ellipse. The engineers at the meeting kept telling them they needn't worry, that MSL was a robust platform that would have no problem accessing terrain outside the ellipse.

The first landing site selection meeting for Mars 2020 is now scheduled for May 12-16, 2014, in the Washington, DC area. I expect the question of go-to versus not go-to sites to come up early on, and it may carry greater weight with the 2020 mission landing site selection than it did with MSL. Remember also that MRO had only just arrived at Mars when they started the landing site selection process last time around; they'll have much more information to use to pick a landing site this time around. It would be great if they could find a compelling science site that requires less driving than Gale does.

The issue of go-to sites is doubly important for the 2020 rover. If one is selected, it would need to go to the site and then return to a safe landing area with the sample cache. Curiosity, on the other hand, can just keep climbing the mountain instead of having to return to the smooth crater floor.

Issues like this makes the SDT's recommendation for more precise landing capabilities more important. If you can land in a tiny flat area near really interesting areas, it's easy to go and return.

My understanding is that current flight certified computers/radiation hardened CPU (i.e. what is on MSL) lags far behind our smart phones and certainly behind our lap and desktop intel Core i-7 processors. Doug and other can answer more accurately, but my sense is that current autonav software probably already taxes the onboard computer capabilities. You are not going to get 10X leap in autonomy using the current computer hardware.

The RAD750 is pretty ubiquitous in planetary spaceflight now. MRO, Kepler, LRO, WISE, SDO, Juno, Curiosity. I would expect 2020 to fly with the same hardware.

The only step-change you might see is if auto-nav processing is handed off to an FPGA configured to do the job. The increase in performance would probably mean there would be a negligible difference between driving with autonav on, compared to blind-drives. However - there is a risk, cost, power etc etc associated with flying such a development.

If they do go with an MSL clone MMRTG and they don't take a 26 month launch deferment surely there'll be a little more power available than MSL had at landing?

A negligible amount that could be entirely swallowed ( and then some ) by a less equatorial landing site and the resultant heating requirements etc.

Really? The RAD750 was developed in 2001 and deployed in 2005. It's been ubiquitous for the last decade, and has performed very well, which counts for a lot. But are there no plans to produce a next-generation chip in the next five years?


Doug M.

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. So I'm sure its possible to develop a follow-on to this hardware (and lots of more) but you have to make choices.

I've not seen an accounting of what is new and what is old, but I believe that most of the new investment is going into an improved science package. Infrastructure (landing, mobility, power, imaging, etc.) are maintained from MSL.

I see that. Developing a chip that's vacuum-tolerant, radiation resistant, and able to function over a range of more than 100 degrees Kelvin is no small or inexpensive thing. And you can reasonably ask how much processing power your average space probe really needs. While more is always nice, right now it seems a luxury rather than a necessity.

But the RAD750 came just five years after the RAD6000 (which is also still widely in use). That was a decade ago, and there hasn't been a new chip since. Googling, I find a 2012 article that says , "Looking down the road to future spacecraft, BAE Systems is developing a quad-core processor for space-based applications that will run at gigahertz speeds... The new computers will be especially useful for image processing." BAE Systems website has a lot of information on space electronics, but AFAICT it doesn't look like a replacement for the RAD750 is in the works yet. Does anyone know more?



Doug M.

Apparently not inexpensive: according to this article http://en.wikipedia.org/wiki/RAD750 we are talking ~$200,000 for one unit. If you happen to have some extras laying around, I'm sure there is a temptation to use them rather than buy the latest and greatest (assuming one exists).

BAE Systems' has a new next-generation processor family. The RAD5545 multi-core processor is the first member of the new RAD5500 family--it will provide a 10X performance increase over the heritage RAD750 according to this PDF: PDF link
EDIT: This posted in response to question regarding next gen BAE family. As indicated below this will not be on the 2020. The RAD5500 will probably start showing up in new spacecraft designs over the coming years.

Go back and read the 2020 AO PIP and observe all the stuff about Technological Readiness Levels.
http://soma.larc.nasa.gov/mars2020/programlibrary.html

Every rover system is going to be a build-to-print copy of MSL unless there is a really good reason to change.

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

Mike has pointed people to the 2020 proposal documentation - one of the best documents to read is this.

http://soma.larc.nasa.gov/mars2020/pdf_fil...8Signatures.pdf

Explicitly...
"The MSL heritage design includes Rover Compute Elements (RCEs) and Rover Power/Analog Modules (RPAMs). "

Same computer. Period.

Note some of the level of MSL reuse - down to using an identical flex cable to go up the mast to mast mounted instruments, for example. The power budgets offered are the same. The thermal conditions the same. Data volumes the same. The available peak current for instruments is the same. Do a search for the word heritage - it appears 50 times. MSL appears 61 times.

I got a drawing for which place you are allowed to present a new instrument for 2020 and this are the "boxes" were the MSL instruments are located. There is big box in the rover, the size of SAM, an ideal place to put the machinery to produce water and oxygen from Martian raw materials. There are several boxes on the arm which are indicated to have a larger temperature scale to tolerate.

All I see there is a 100% copy of MSL with other instruments, but only if they have the same volume, energy consumption and mounting points like the old ones.

For an team to built an instrument a challenge, but a manageable one.

In the QA to the instrument AO, NASA said they would consider instruments mounted to the outside of the rover body on a case by case basis. So some additional volume might be possible. If any proposals like this were made, they might be like the Phoenix top-of-lander instrument boxes.