MSL Approach Phase Options

[Tesheiner]

[Pertinax]

Hi Rob,

Thank you for you post, particularly for your wonderfully visual description of methodically working through a seemingly impossible problem.

-- Pertinax

[Guest_Oersted_*]

I wonder whether the planning and programming for non-nominal landing scenarios includes alternative thrusting profiles in case the bridles snag or are blocked as they spool out. Say, if the rover only descended half a meter on the bridle, would the descent stage then be able to recognise the situation and descend accordingly? It would of course mean a higher risk of plume damage, but better a dinged rover than no rover.

[MarsEngineer]

I wonder whether the planning and programming for non-nominal landing scenarios includes alternative thrusting profiles in case the bridles snag or are blocked as they spool out. Say, if the rover only descended half a meter on the bridle, would the descent stage then be able to recognise the situation and descend accordingly? It would of course mean a higher risk of plume damage, but better a dinged rover than no rover.

Hi Oersted,

In our "stress and robustness" simulations (which we have been doing a lot of in the last few months), we try to think up cases like this to see where the system "breaks" using one of our software simulations (we have a couple of variants of simulation for different purposes - plus the testbed with all of the electronics in it too). We have tried cases like: slower (sticky) engine throttle responses, lower or asymmetric throttle engine performance, super slow bridle deployment. I don't think we tried the case where the bridle does not fully deploy (I will ask), but I am pretty sure it would work to a point - the architecture does not depend on the bridle length being "just right". The software doesn't "recognize the situation" per se, the software really does not monitor nor really care about the position nor orientation of the rover with respect to the descent stage, it simply waits for he ground to show up and "off load" the descent stage. The software "knows" that the rover is on the ground when it notices that the throttle settings have cut in half (as it must in order to maintain the descent rate of around 0.7 m/s even after the rover is on the ground - at least for a couple of seconds). And for that reason, a half meter bridle would mean that the descent stage would come down and make contact with the rover in those 2-ish seconds that it take for the software to notice that the rover has been off-loaded (remember there are uptake reels on both ends of the bridles to take up the slack). Plus there would be a lot more plume issues being that close to the ground.

I have personally witnessed the BUD (bridle/umbilical device) testing as well as the system level separation tests and I know the design pretty well. John, Tom and the gang put a lot of thought put into it as well as incorporating lessons from MPF and MER (which used centripetal friction brakes instead of an electro-magnetic brake). You might recall that on Spirit landing, the bridle's brake was a tad "too good" (too much friction) and the lander with Spirit inside took "much" (a few seconds) longer than we expected to deploy the bridle to full length (over the coming 3 weeks before Opportunity arrived, we frantically re-re-tested the brake in the expensive 25 ft space simulator). This new design avoids this possibility. It is a nice clean and elegant design they came up with. This does not keep me up at night.

If and when we find stress cases that do not work, we decide whether or not to do anything about them depending on our best judgement of the probability of occurrence vs level of difficulty in mitigating the threat. More often than not we find that that system muscles through the problem (for example a very short data outage on the bridle) , however when we consider "hard failures" (such as a failure to cut a bridle) we already know the outcome will not be good and we have known that from the beginning. In those cases we count on the reliability and simplicity of he design ... and of course we test it like crazy to "prove" it (we often misuse the word "proof" which implies deductive reasoning, rather we should say "demonstrate" for the inductive reasoning behind the bulk of our reliability arguments).

(I am really procrastinating today. I love to talk about this stuff.)

-Rob

Opinions expressed are those of the author and do not necessarily represent the views of JPL/Caltech/NASA.

[nprev]

Rob, if you have the time & feel like it, I'd love to hear the history/thinking that led to the descent stage flyaway parameters.

When I first heard about it I figured the easy way would be to tilt it about 5 deg after bridle separation in any direction & let it rip full throttle until it smashed into something or ran out of fuel & crashed. Understand that the flight path is a bit more "elegant" now.

[Guest_Oersted_*]

......

(I am really procrastinating today. I love to talk about this stuff.)

-Rob

Rob, thanks for your informative and very interesting reply. We're fortunate to have caught you during those few extra moments when the preparations for the party are done, the rehearsals are all over, and the guests and relatives are just standing around, chomping peanuts and having a drink while awaiting the bridal, sorry, bridle arrival... ;-)

We know that you are of course actually very busy, but that just makes it an even bigger privilege to get these insights.

[DavidVicari]

Rob,

Thanks for the great replies!!

You said the bridle doesn't keep you up at night... I have to ask, is there anything about the landing that is keeping you up??

From an outsider the thing that worries me the most is the parachute deployment. It seems like the least controllable and most chaotic of all the events. My second biggest concern is the number of pyros that must work. I know that they are extremely reliable but it still scares me a little.

[mchan]

Another thanks to MarsEngineer. Your perspective on the engineering challenges give outsiders a view of the work on the inside, and are always informative and a pleasure to read.

Regarding the parachute, there is a great series of videos on the design and tests of the parachute and of the folks doing the design / test. You can see how early tests resulted in shredded chutes. One video featured a great practical joke where the data acquistion appeared to be bad in a full-scale wind tunnel test. This aspect of EDL appears to be well tested.

[Guest_Oersted_*]

"JPLnews" Youtube channel:

Martian Series: Testing Curiosity's Parachute Part 1:
http://www.youtube.com/watch?v=O7vf2HUMMdo

Martian Series: Testing Curiosity's Parachute Part 2:
http://www.youtube.com/watch?v=JRRcbZlofOk

Martian Series: Testing Curiosity's Parachute Part 3:
http://www.youtube.com/watch?v=-NJamPhtRjA

Martian Series: Testing Curiosity's Parachute Part 3:
http://www.youtube.com/watch?v=J6TceTZq1L0

[claurel]

I noticed that there's a relatively recent SPICE kernel available for MRO that covers the time window around MSL's entry, descent, and landing. I was curious what view MRO would have of Gale Crater, so I created a short visualization of the two spacecraft's trajectories near the time of the landing.

http://youtu.be/-f0BDnJNW-8

MRO's trajectory is plotted in a Mars-centered inertial frame, while MSL's is depicted in a Mars-fixed frame. Of course, the actual EDL trajectory of MSL is likely to vary somewhat from the predicted one in the SPICE kernel.

--Chris

[fredk]

Sorry if this has been discussed, but I noticed Scott Maxwell say in a recent interview:

With MSL about to land and join his older sister on the red planet...

We've always referred to the MERs as "she's". Is the team referring to MSL as "he"? Oppy's younger (but bigger!) brother?

[elakdawalla]

Scott insists that the rover is a "he" and affectionately calls him "George." I'm not sure of the origin of that name -- I'm sure he'd tell me if I asked him! -- but I always imagine the Looney Tunes yeti whenever I hear the name, which may itself be a reference to Of Mice and Men. (Of course the yeti's name wasn't George, that was the yeti's name for his pet bunny rabbit, but still...)

I, however, think Curiosity is a girl. So there.

I've heard "he," "she," and "it" from various people involved in the mission. Gender is, I guess, in the imagination of the speaker. You'll never see an official JPL press release refer to the rover as anything but "it."

[stevesliva]

^ Curious George

[djellison]

I call the testbed rover George - but I'm still getting used to Curiosity for the rover - and vessels of exploration are always she's - that's something Scott and I just have to disagree on

[RoverDriver]

Since curiosity in italian is LA curiosita' (femminine), MSL is definitely a she.