Help - Search - Members - Calendar
Full Version: Skycrane
Unmanned Spaceflight.com > Mars & Missions > MSL
Pages: 1, 2, 3, 4, 5, 6
Stephen
QUOTE (djellison @ May 2 2005, 09:00 AM)
QUOTE (Stephen @ May 2 2005, 01:25 AM)
What about marrying a Montgolfier-type balloon with a small rover? The balloon does most of the travelling, snapping pics etc from aloft, but lands at interesting sites to release a small (Sojourner-style?) rover. The rover does its business (with the balloon's own  instrument package acting as a basestation Pathfinder-style), then returns, gets back onboard its cradle (or whatever), and off they fly to another site.

How do you stop the balloon just drifting away in the mean time?

Good question! I'll let the engineers figure that one out. smile.gif

Actually, what you'd presumably need would be some sort of anchoring system. Something that would cling to or drill into the martian surface, perhaps; possibly in conjunction with a temporary partial deflation of the balloon.

Alternately you might use something akin to pitons. Climbers use those to secure themselves to rockfaces. Something similar might be used to secure a balloon to the surface of Mars (fired into the ground maybe by some kind of small explosive device?), although unless they were reuseable that would obviously limit the number of times the balloon could anchor itself. On the other hand, the pitons might also double as sensor-carriers, similar perhaps to the ones that were piggy-backing on the MPL.
tedstryk
QUOTE (djellison @ Apr 30 2005, 06:01 PM)
QUOTE (Bubbinski @ Apr 30 2005, 05:12 PM)
I thought to myself when I saw the pics of the MER landers that they should have had a few basic instruments themselves to at least act as long term "weather stations" instead of just being lumps of metal.  But maybe weight/funding restrictions prevented this?


Well - you'd have to have a power source, a data handling system, instrumentation, comms packages, in a place with no mass budget, no volume.....

Doug
*



The only way I could see this work is to have a simple camera, a small solar panel, and a small receiver on the rover, a la Pathfinder/Sojourner, to get the images back. Still seem like it isn't worth it considering the risks involved with the additional complexity.
helvick
QUOTE (Stephen @ May 2 2005, 12:28 PM)
Alternately you might use something akin to pitons. Climbers use those to secure themselves to rockfaces. Something similar might be used to secure a balloon to the surface of Mars (fired into the ground maybe by some kind of small explosive device?), although unless they were reuseable that would obviously limit the number of times the balloon could anchor itself. On the other hand, the pitons might also double as sensor-carriers, similar perhaps to the ones that were piggy-backing on the MPL.
*


My gut feeling is that the trade offs in terms of ballon assembly mass vs the mass of any rover would be very wasteful. However this website says that one potentially workable design has a 65kg balloon assembly, 15kg gondola and a 13.5kg instrumentation "Tail rope". That is actually enough to carry a Sojourner size rover with a couple of kilos to spare.

As far as anchoring is concerned - rather than pitons something like an SLCD combined with a drilling unit should be able to hold a ballon against it's lifting forces easily enough but it would have serious trouble with Martian winds. The balloon described in the link above would be ~27m in diameter - that's a very big sail even in an atmosphere only 1% as dense as earth's. With a gentle (by earth standards) force 3 breeze that would generate around 100N of force probably equivalent to the peak lifting force of the balloon but with a typical Martian wind (thought to be 200kmh) it would have to hold 10kN, and in a peak Martian gust (600kmh) that would be 93kN. To put that in perspective that's about the same as the peak output from a medium sized family car, well here in Europe anyway.

Small balloons relatively high in the atmosphere with very specific lightweight meteorological sensors might be worthwhile on Mars but that atmosphere is a bad place to try something like this.

The Tumbleweed Rover, mad and all as the idea seems, looks like a good way to harness some ambient free energy to get sensors distributed over long distances from an initial landing site. The target mass of a Martian equivalent would be about 40kg so you'd get 2 for the price of the single balloon, they are much simpler and if the Antarctic tests are anything to go by you could expect to get 70-100km range without much trouble.
ilbasso
From the JPL Tumbleweed Rover website cited above:
"The Tumbleweed Rover is a large, inflated ball that can be windblown and used to explore the surfaces of Earth, Mars, Venus, Titan, and perhaps Saturn’s moon Io (via supersonic volcanic wind) and Neptune’s moon Triton (as evidenced by significant surface wind erosion)."

Looks like Io has moved!! Why weren't we informed?????
lyford
Y'all can come party with me over in The Tumbleweed Thread also on this great board!

(Funny how this is not the first time a discussion of Skycrane has tempted folks to speak of our wind blown friends...) tongue.gif
Cugel
...0. (beaming down from the other thread)

Marcel,

Well, it's not so easy as I proposed of course. However, you didn't completely convince me, yet! For instance, why couldn't a fixed decent stage put the rover down with an arbitrary (almost zero) low speed? At least the landing velocity would not have to be greater than when its dangling from a cable. The suspension system will have to be fixated during landing, that is true. But the six (inflated) wheels will provide some shock absorbing, at least.

As far as dangling on a rope is concerned, I don't see how you can prevent it from swinging too much.

Greetings from Bilthoven!

Henk
BruceMoomaw
I asked MSL's project manager about precisely this question at the first meeting of NASA's Mars Strategic Roadmap Committee in January. He told me that the force put on the rover's undercarriage by a descent stage to which it was rigidly fastened -- even if one tried to null out as much of the descent stage's speed as possible -- would be sufficient to endanger the rover, and that for this reason they do indeed prefer dangling it from a cable.

However, he also emphasized that it's very much up in the air as to whether a Skycrane control system CAN be developed to keep dangerous swinging from occurring -- and that the tests in the Mojave Desert this summer will be crucial in answering that question. If it can't, they'll fall back on a "pallet lander" design, in which the rover sits on top of the descent stage which simply plumps down directly onto the Martian surface on its belly, using the crushing of its own descent rockets and fuel tanks as a shock absorber, after which the rover climbs off the top of the descent stage using a ramp. This of course does involve some risks -- there just might be an minor explosion in the descent rocket system, and certainly there's a danger of some of its hydrazine fuel being squirted around and contaminating the immediate landing site -- and the added weight of the rover's ramp is a problem. But if Skycrane desn't work, it's the only solution they have left that has a mass anywhere near acceptably low enough for the mission to be flown.
Marcel
Bruce,

I don't understand. How could they be so sure that a stage on the back of MSL could endanger the integrity of the wheels/mobility system due to it's weight, while they are so confident that EXPLODING hydrazine won't hurt MSL. The latter SEEMS like a very bad idea to me. If the skycrane concept doesn't work and they'll go for the pallet, they better integrate airbags (or suspended legs, which have been discussed earlier in this thread) underneath to absorb the shock, instead of letting the tanks and rockets crush near a billion dollar vehicle.

Isn't it ?

Cugel: As far as my knowledge goes considering the next wheels on Mars, i don't think they're going to be inflatable. There's no reliable flexible material that will last for years in the Martian environment. Surely the have to be bigger than MER wheels, but inflatable.....besides: who's going to repair them if they're flat ? laugh.gif About your other point: the descend velocity: there's always a margin of error AND a timelag between what the rover actually sees as it's momentairily (?) status (velocity, angular movement, terrain aspects) and the next needed correction in order to get to it's goal: the surface. There will always be a considerable bandwith of vertical speed that has to be taken into account. Besides: landing at 0 km/h isn't a good idea. It's like landing a plane. You don't want to do it as soft as possible, because the chance of instable behaviour (you're not in the air, but you're also not on the ground yet) like bouncing, treamendous increasing wind vulnerability, etc. It has to be a gentle drop from, let's say 50 cm to a meter(not "fwwwp", not "BOOOM", but "thump").

Something else I wonder: Imagine a palletlander hitting a rock like Yogi with one of it's corners while stilll hovering on the other 3 engines: i'll eat my shoe if someone can devellop an autonomous descend stage that can actually lift the whole thing again, hover sideways some meters and put it down again. You'll need a Neil Armstrongish kind of autonomy to deal with that....how are they going to recognize these threats autonomously ? Can this be done reliably ? I hope the skycrane tests will work out fine.....because a pallet lander has this risk, that a skycrane does not.....
Cugel
Gentlemen,

Thanks for the response(s). To wrap it up:

skycrane's pros
1. The weight of the descent stage doesn't have to be carried by the rover wheels at touch-down.
2. Lowering the rover on a cable gives more margin (for error) in the vertical range.
3. The previous points put together make a dedicated landing gear superfluous.
4. Which means you're mobile, right after landing....

skycrane's cons
1. Dangling on a single cable means there is no control in the horizontal range. (It can swing and spin).
2. It adds a mission critical mechanism. (and a rather complicated one!)

To be honest, I would be amazed if they can turn this idea into reality. But it would be cool.
BruceMoomaw
I don't yet fully understand their reasoning on the Pallet Lander myself -- it may be that it will have so little fuel left that the risk will be minimal. (They might even have it hover a short distance up unti its fuel is completely exhausted.) As for the danger of it tipping over, though: MSL will have, in any case, a sophisticated landing-obstacle avoidance system using phased-array radar to locate and avoid any big boulders or dangerous slopes before touchdown, and the Pallet Lander designs also call for a set of short outrigger poles sticking out from the sides to prevent tipover.
djellison
I dont think that lowering the rover down is any more challenging that lowering, say, the lander from the backshell for the MPF/MER edl sequence. It's a bridle with a rate limiter.

If you have that system, but upside down (one cable out of skycrane, down to a shackle, with then 3 cables going to front, back right and back left of the rover ) it's actually fairly simple.

And you have to admit - compared to making the whole thing landable, and having egress aids etc etc - a bridle and a limiter are a LOT lighter.

The parachute/backshell/airbag/DIMES/TIRS system was just insane - skycrane is no worse imho. It presents different challenges - but if you can get a lander to put itself on the ground just using rockets - you can get it to put itself a few M above the ground with rockets and drop the thing on a rope. Yes - it'll impart some force on the crane, but we're talking about a couple of seconds.

Bruce - any idea when things are going to get going with the testing out in the desert?

Doug
Cugel
Doug,

Actually, I completely forgot about the MER system! However, the MER backshell lowered the rover+balloons into mid air. The MSL descent stage must lower the rover to solid ground. IMO this puts serious limits on the lateral movement of the rover at touch down. This requirement will result in a complex system of sensors, thrusters and software that has to cope with unpredictable wind conditions... Well, for the guys at JPL it's probably just another day at the office!

p.s. If the skycrane doesn't work out, we can always fedex the darn thing....
djellison
I was thinking of this - but the other way around smile.gif



Doug
dvandorn
I think perhaps the operation of the backshell retros on the MERs and on Pathfinder gave the JPL guys the idea of the Skycrane, in that in each case, the descent rate of the vehicle was virtually arrested and the lander package simply dropped from a safe height in each case. If we can achieve a zero descent rate at a given point above the surface, why not try to achieve it *at* the surface with a similar system?

The problem, of course, is that in each of the three airbag landings, the lateral velocity was significant -- enough to tear up MSL with no airbags. I think, though, that the dynamics of the MERs' descents show that, after the deployment of the lanyard holding the lander, the backshell is what oscillated, the parachute and the lander stayed relatively well in place. Since the backshell was oscillating in a small circle around the line between the lander and the parachute, retro fire occurred at a random angle to the descent vector. I think that's what tended to impart the lateral velocity.

If the Skycrane separates from the parachute and begins to descend on rocket power *before* MSL is belayed down its lanyard, the natural dynamics of the descending sytem will keep the random oscillation of the rocket platform from occurring, as it has on the airbag landings. Which means the main cause of lateral velocity could be removed from the system, making Skycrane viable. The problem, of course, is that additional lateral maneuvering by the Skycrane after lanyard deployment can cause MSL to swing laterally, thereby spelling disaster.

I think the best chance Skycrane has of working is if it finds its intended landing point, maneuvers to a point directly above it, starts a completely vertical descent (with no lateral velocity *at all*) to that point, and *then* belays the MSL down its lanyard. As long as you drop the lanyard while you're in a completely vertical descent, and as long as the weight on the lanyard is balanced properly, then wind is the only factor that can possibly induce lateral motion, and hopefully, unless the wind is *really* blowing hard, odds are good the final descent can proceed with little to no lateral motions developing.

But if we use the Pathfinder/MER deployment sequence, which would put the Skycrane module in the middle of the system and thereby ensure that will oscillate around the line between the parachute and the MSL, at parachute separation and rocket ignition you will almost definitely have a swinging motion going between Skycrane and MSL which will be very hard to damp down to acceptable levels before you reach the surface. So it seems to me that Skycrane can work, if it uses the proper deployment sequence.

-the other Doug
Jeff7
Some have suggested a "pallet" for the MSL. I am not sure if it's been suggested, but what about a hybrid? A pyramid shaped cocoon, but without the bulk of the airbags. It wouldn't need to have all the associated equipment either - just a basic system to be able to open up. That would solve the problem of risking damage to the rover's wheels from a hard landing or swaying, as well as the issue of setting a pallet down on a large rock and tipping the whole thing over.
If the rover is in the pyramid thing, it could land on a boulder while swinging around, and still be able to right itself.
And as for jarring around, I think that they said the MER's endured 40G's when they bounced down. I don't think that a light tumble from a big rock onto sandy terrain would impart nearly as much force into the MSL.
djellison
QUOTE (dvandorn @ Jul 8 2005, 08:05 PM)
If we can achieve a zero descent rate at a given point above the surface, why not try to achieve it *at* the surface with a similar system?


You're going to cake the entire spacecraft in dust, and ask the rocker-bogie to deal with the whole lot at some point as well.

I think skycrane is easily dismissed as mad - but so was the Pathfinder plan smile.gif

Doug
dvandorn
QUOTE (Jeff7 @ Jul 8 2005, 03:38 PM)
...as for jarring around, I think that they said the MER's endured 40G's when they bounced down. I don't think that a light tumble from a big rock onto sandy terrain would impart nearly as much force into the MSL.
*

The MERs were literally bolted down into their folded positions when the landers bounced in. And much of the shock was attenuated by the pyramidal shell itself.

If you subjected the MERs to a 40G jolt now, in their deployed configurations, you'd wreck 'em, for sure. Depending on the vector of such a jolt, the rocker-bogie suspension system would break, solar panel connections would break, the camera mast would break -- you'd lose the vehicles, most likely.

And remember, MSL is to be dropped onto the surface, wheels-down, in its fully deployed configuration. Carrying a heavy radioisotopic thermal generator on its back. Touchdown will have to occur at something like .1 to .3 meters per second, with *no* significant lateral velocity, for this to work.

-the other Doug
Jeff7
QUOTE (dvandorn @ Jul 8 2005, 05:43 PM)
The MERs were literally bolted down into their folded positions when the landers bounced in.  And much of the shock was attenuated by the pyramidal shell itself.

If you subjected the MERs to a 40G jolt now, in their deployed configurations, you'd wreck 'em, for sure.  Depending on the vector of such a jolt, the rocker-bogie suspension system would break, solar panel connections would break, the camera mast would break  -- you'd lose the vehicles, most likely.

And remember, MSL is to be dropped onto the surface, wheels-down, in its fully deployed configuration.  Carrying a heavy radioisotopic thermal generator on its back.  Touchdown will have to occur at something like .1 to .3 meters per second, with *no* significant lateral velocity, for this to work.

-the other Doug
*



Ok, so bolt MSL to the inside of the shell. Maybe line the pyramid with some Aerogel to absorb some of the impact force.
This pyramid system would definitely solve the problem of setting the rover down on a rock. Plus, one other thought - ok, so the Skycrane sees where the ground is, and as the rover repels down, it knows when to stop. But will it be able to see a rock? What I'm wondering here is, not so much will it get stuck, but will it still be travelling too fast. The rover thinks the ground is still a half meter away, so it doesn't slow down in time, and it smacks its wheel hard on a rock that it didn't know was there. Or it punches a hole right in its underside.

Yes, it would add some weight. But it wouldn't need airbags, and aerogel is very lightweight. It could conceivably protect against the scenario I described, as well as a few others detailed in this thread thus far.
djellison
QUOTE (Jeff7 @ Jul 9 2005, 03:21 PM)
Yes, it would add some weight.


And in doing so - make MSL unfeasable.

Doug
Marcel
QUOTE (djellison @ Jul 8 2005, 08:43 AM)
I dont think that lowering the rover down is any more challenging than lowering, say, the lander from the backshell for the MPF/MER edl sequence.
Doug
*


I'd say it's actually easier from an engineering point of view ! And lighter: you don't have to bolt together a cocoon around the rover, introduce extra pyro's (that actually CAN fail), wrap it in a heavy spacious airbag-system, changes of ramps stuck in pieces of airbag....

The only difference (which makes it more complex from an avionics/detection-system point of view) is, that it has to be a hell of a lot more accurate than the way MER were landed. OK, let's say recognizing hazardous terrain can be done reliable. The craft then has to halt completely. And stay there to lower the rover. And these last two are not so easy, considering windshear.

Can this movement be corrected using lateral rocket engines ? Maybe like the "Harrier" kind of steerabillity (a main engine in the heart, 4 to 6 steerable, more or less vertical exhaust nozzles that can compensate for horizontal drift ? The only thing you need then is a system that can compare "airspeed"and "ground speed", but that should be no problem, compared to a system that can evaluate hazards in the terrain autonomously ! "Just" add another function to that laugh.gif
RNeuhaus
Another tought. As the MSL will be capable to run up to 50 km per day. It would be a good idea to land a very flat and easy land and where there is no tracks of dust devil (calm zone). Once it is landed, the MSL can get a desired point in few soles. For simplicity, I prefer the QUOTE(dvandorn @ Jul 8 2005, 08:05 PM)

"If we can achieve a zero descent rate at a given point above the surface, why not try to achieve it *at* the surface with a similar system?"

The reason is the simplicyty and reliability of operation with not have much avionics, control, electronic to control the descending against the wind, rope, etc. Like the same way as did the Vikings: Long landing strong arms.

The drawback which says QUOTE(djellison@ Jul 8 2005, 05:15 PM)
"You're going to cake the entire spacecraft in dust, and ask the rocker-bogie to deal with the whole lot at some point as well."

It is of no matter. The MSL must be very well covered and protected agains any dust like a flower.

Rodolfo
Marcel
Oops, sorry, button was pushed before i knew, accidentally replying with just a quote.

Anyway: I do NOT agree with the statement that DD free zones are safer to land than other places. Wind gusts are everywhere on Mars and a craft solely depending on rocket engines cannot be made simple. I agree that they need to be as simple as possible, but it is not as simple as you said by the quote.

The fact is, that at a certain point above the surface, both MER's were brought to a halt, yes, but it was not controlled in a way that an airbag free craft requires. In fact, you NEED to control the point where it comes to a complete stop and you have to STAY there (up in the air) for a while. The MER's EDL system was not designed for doing that and it needs a completely different system to just make that point coincide with the surface. THAT's the whole point. That is not a simple thing to design. It's hard. Especially because you want it to work. For sure. It needs to be robust, it needs to be reproducable during testing, again and again.

Considering your point with respect to covering MSL for dust: the whole point why you want to use a skycrane, is that

1: There's no risk of malfunction during an egress sequence, simply because there isn't any,
AND
2: You can start doing science right away
AND
3: You don't need to make a heavy cocoon around the rover. Adds up to the payload. Waste of valuable science producing kilograms to wrap the whole thing just to prevent dust entering the vehicle. And adding up risk of getting stuck in the sequence that has to get rid of the cover.

I'd say: make it dust proof, just like MER (except for the solar panels) by encapsulating only the dust critical components (lenses, bearings, instrument arm, etc).

And about the 50 km's of driving in a day ohmy.gif ohmy.gif : let's take some steps back, with Oppy's adventure freshly in our memories unsure.gif !!

Marcel.
djellison
QUOTE (RNeuhaus @ Jul 11 2005, 02:37 PM)
As the MSL will be capable to run up to 50 km per day


blink.gif

I've not heard any figures that large. Perhaps 1% of that, but certainly not multiple km's per sol

Doug
RNeuhaus
QUOTE (Marcel @ Jul 11 2005, 10:12 AM)
Oops, sorry, button was pushed before i knew, accidentally replying with just a quote.

Anyway: I do NOT agree with the statement that DD free zones are safer to land than other places. Wind gusts are everywhere on Mars and a craft solely depending on rocket engines cannot be made simple. I agree that they need to be as simple as possible, but it is not as simple as you said by the quote.
*


I tought that the wind gust will occur mostly around the Equatorial line and summer times and not all other places.

QUOTE (Marcel @ Jul 11 2005, 10:12 AM)
The fact is, that at a certain point above the surface, both MER's were brought to a halt, yes, but it was not controlled in a way that an airbag free craft requires. In fact, you NEED to control the point where it comes to a complete stop and you have to STAY there (up in the air) for a while. The MER's EDL system was not designed for doing that and it needs a completely different system to just make that point coincide with the surface. THAT's the whole point. That is not a simple thing to design. It's hard. Especially because you want it to work. For sure. It needs to be robust, it needs to be reproducable during testing, again and again.


I was not well informed. I have seen the video examples on how the MER will land on Mars after the parachute is stood quietly in vertical position. But, now I see that this is not true at all.

QUOTE (Marcel @ Jul 11 2005, 10:12 AM)
Considering your point with respect to covering MSL for dust: the whole point why you want to use a skycrane, is that

1: There's no risk of malfunction during an egress sequence, simply because there isn't any,
AND
2: You can start doing science right away
AND
3: You don't need to make a heavy cocoon around the rover. Adds up to the payload. Waste of valuable science producing kilograms to wrap the whole thing just to prevent dust entering the vehicle. And adding up risk of getting stuck in the sequence that has to get rid of the cover.

I'd say: make it dust proof, just like MER (except for the solar panels) by encapsulating only the dust critical components (lenses, bearings, instrument arm, etc).

I tought that Scycrane will add more weight to landing capsule but I realized it is no so, and it would be about the same as the ones like of Vikings.

Scycrane with wings, the things would be worst (greater power to break) and I agree that the Scycrane must land from a vertical position. What happens is that I have not seen a unmanned vehicle to land vertically likes the helicopters and Harris. I assume that to control the vertical landing requires greater control than wing landing. One thing that will help much in landing control on Mars is to its lower gravity.

QUOTE (Marcel @ Jul 11 2005, 10:12 AM)
And about the 50 km's of driving in a day  ohmy.gif  ohmy.gif : let's take some steps back, with Oppy's adventure freshly in our memories  unsure.gif !!


I am sorry. huh.gif My memory has failed more or less. I knew that MSL is by far more capable than MER that can travel no more than 200 meters per day and MSL can ....

http://www.space.com/businesstechnology/te...lab_050105.html

"Theisinger explained. With an anticipated mobility range of some 12 miles (20 kilometers), "we expect it to last two Earth years…one Mars year," he added."

Rodolfo
odave
...interesting that the MERs are closing in on one Mars year, which is- ~668 sols, IIRC. Here's hoping that MSL's actual lifetime scales with the estimate the same way smile.gif
BruceMoomaw
Actually, Mars' air is so thin that its winds -- even at their fastest -- are absolutely no problem for a vehicle that does NOT use a parachute all the way down to the near-surface. One of the main problems with the Pathfinder/MER airbag system, however, is the fact that horizontal winds obviously ARE going to be a severe problem with a lander that remains dangling from its very big parachute until it's just a short distance up. Indeed, if it hadn't been for the frantic last-minute effort to install the Rube Goldberg-like DIMES system on the MERs to measure and cancel out a large part of their horizontal drift just before landing, there is a good chance that Spirit would have hit the ground at such a high horizontal velocity that its airbags would have ruptured and we would have lost it.

But -- to repeat -- the concerns with Skycrane, though serious, do not involve winds; even the strongest winds on Mars are unlikely to cause any serious cable-swinging or other control problems. The problems, instead, involve the inevitable sloppiness and errors in the lander's own attitude and position control systems.
Marcel
QUOTE (BruceMoomaw @ Jul 12 2005, 01:37 AM)
Actually, Mars' air is so thin that its winds -- even at their fastest -- are absolutely no problem for a vehicle that does NOT use a parachute all the way down to the near-surface. 
*


Bruce,

Then it should NOT lower the rover on a rope until rocket power took over completely and the parachute is jettisonned. Am i right that that's the best way to do it ? Then why was there concern (also in this thread) about the dangling ?

Let me see if i got the picture right about the plans as they are now:

Entry,
parachute,
jettison heatshield,
lower skycrane/MSL as a whole on a tether, away from backshell
startup engines of skycrane
jettison parachute, backshell and tether
lower skycrane/MSL until stable hovering above surface within, let's say 20 meters
THEN maintain skycrane at same altitude and
lower MSL from the crane
detach MSL/skycrane at the moment of ground contact
remove skycrane elsewhere.

Where in this sequence was the fear of dangling/turning ? I don't see the problem....

The only problem around dangling i can see, is if you let MSL hang on the skycrane, while the parachute is still on top. But that's not how it's going to be, is it ?

SO, "no problemo", except the aspect of instable behaviour due to inertia and oscillation/resonance (overcompensation, runaway behaviour). That can be tackled i'd say...Viking did it (twice), why not MSL. wink.gif
djellison
I dont believe there's any skycrane-backshell tether. It would just jetison the backshell like Viking/Phoenix

Doug
Marcel
QUOTE (djellison @ Jul 12 2005, 08:17 AM)
I dont believe there's any skycrane-backshell tether. It would just jetison the backshell like Viking/Phoenix

Doug
*

And how are they going to make sure that the whole craft is oriented properly (instead of tumbling without a parachute) before rockets are activated ? Or will the backshell/parachute be jetisoned WHILE rockets already burn ? Don't understand then how to prevent the parachute from draping over the craft....
edstrick
There may be small "jettison" rockets on the chute lines where they're attached to the skycrane, or under the aeroshell if that separates from the skycrane.

Or the skycrane may fire it's rockets at "stabilizing" thrust levels for 5 or so seconds as it drops away from the chute before going to full brakeing thrust.
djellison
http://phoenix.lpl.arizona.edu/multimedia/...nimation_hq.mov

Basically - I think it's like that - but the whole thing will come to a dead stop above the ground and do the skycrane 'thing'

You could always drop out the backshell whilst still on the 'chute I guess - that might eliminate an opportunity for instability

Doug
RNeuhaus
QUOTE (Marcel @ Jul 12 2005, 02:08 AM)
Bruce,
lower skycrane/MSL until stable hovering above surface within, let's say 20 meters
*


20 meters is like a building of 6 floors....to long to hang the rope...

http://www.space.com/businesstechnology/te...lab_040211.html
Using guidance and navigation gear, the Mars-bound hardware enters hover mode for a nominal five seconds. In a steady-as-she-goes manner, it hangs in mid-air a mere 15 feet (5 meters) above a pre-determined slice of martian real-estate.

I agree about the Bruce's comments:

1) Mars' wind is not much concern to Skycrane, tether and MSL which has the total would be (180 kg of MSL and the rest might be around of 270 kg -just guess-) that is by far heavier to withstand for a very thin Mars gust winds. It is like when a small car pass a truck. The truck does not feel anything when it passes but the little car do shake up much after passing a truck.

2) About the the inevitable sloppiness and errors in the lander's own attitude and position control systems. It is what I am most concerned. But it can be overcome by lots of testing and perfectioning the electronic with faster microporcessors and RAM to process the radar and images in mapping the landing before selecting the right place to land. I believe, that this problem can be solved and eventually to be perfectly feasible for that mission.

However, anyway, the landing place must be selected like an airport, with minimal security measures such as the flat land, a good distance of free mountains and low horizontal winds (wind shear) and the MSL can drive up much quicker than MER (up to 100 times further) up to the desired place.

One thing that I am concerned of MSL is about the height of mast camera that would be a bit taller than MER. I would like to change the statitic mast for an hydraulic ones likes an submarine's telescope so that the mast can rise taller and be able to map the land with greater distance and help to plan better on the route. This is valid when the MSL drives fast and also help to select the best route over the dunes.

Hope that the Skycrane will improve the landing ellipse error up to 5 by 10 kilometers. For that, first, I would like to see that Skycrane landing technology can drive an helicopter or Harris to land safely on the right place. After that, I would be confident that the mission of MSL will be more predictible.

Rodolfo
djellison
Given the sort of imagery we'll get from MRO - I dont think we really need to be able to look 'further' from the ground.

Doug
RNeuhaus
QUOTE (Marcel @ Jul 12 2005, 04:22 AM)
And how are they going to make sure that the whole craft is oriented properly (instead of tumbling without a parachute) before rockets are activated ? Or will the backshell/parachute be jetisoned WHILE rockets already burn ? Don't understand then how to prevent the parachute from draping over the craft....
*


I think that the parachute won't be able to enclosure the skycrane since after the jettision to break the parachute's rope, it will fly away due to the wind action, a short time after, the skycrane 3 thrusters - one for each 3 legs??- will be fired and start to control the height and position according to the gyroscopy control, to reduce the speed, and to guide to the desired place and lately will fire others 2 or 4?? central thrusters to get the skycrane in hovering an height of 5 meters above of the land before it releases the tether.

One observation, I have seen that the helicopter's pilot spend lots of time and effort to control the position of load attached to a rope. The shorter rope is better, the calm weather (no winds) is better. So I tought that the electronic control for this mission must be of foremost technology. After that, no pilot will fly an helicopter.....

Rodolfo
djellison
Well - a helecopter pilot has cross winds to worry about smile.gif

Not only that - he also has to get the payload down on a specific point.

SKycrane can simply hover in any safe place - it doesnt have to drop MSL onto a big red X on the martian surface

Doug
RNeuhaus
QUOTE (djellison @ Jul 12 2005, 10:57 AM)
Not only that - he also has to get the payload down on a specific point. 

SKycrane can simply hover in any safe place - it doesnt have to drop MSL onto a big red X on the martian surface
*

As I am software specialist, I suppose that the computer software need to see a permited landing with some kind of geometric lines: "O" , square, triangle or rectangle before dropping the MSL.

Let us hope that the computer software won't confuse a purgatory sand with a really nice flat and pavement sand. unsure.gif

Rodolfo
dvandorn
All of the good points made about the stability of the descending MSL "stack" point out why I said that the MSL itself can't be lowered down its rope until after backshell/parachute jettison, and preferably not until the Skycrane is descending vertically or hovering, with no significant lateral velocity.

The problem is, you want to use passive deceleration (i.e., the parachute) as long as possible. The Pathfinder/MER backshell retro-rockets only had to fire a short burst to cancel the vehicle's remaining descent velocity, so they didn't have to carry much fuel and could be pretty lightweight. Skycrane will require something like 30 seconds of sustained rocket firings, at variable thrust, to achieve its 5-meter "hover" and still have enough gas left to fly away, out of range of the MSL so that it doesn't crash back down on top of the rover.

So, it's a balancing act between hanging on to the parachute until the last possible moment, thus saving fuel (and allowing Skycrane to carry enough fuel to achieve its mission), and giving Skycrane enough time and altitude to achieve its 5-meter hover.

And for those who think that the Skycrane approach enhances the chance of mission success by reducing the number of little pyro actuators that all need to work correctly (as the MERs had to do, in the deployment sequence), that *is* true -- but on MSL, you have to rely on the pyros that separate MSL from its bridle, or else Skycrane will wreck the rover as it tries to fly away. So, while you can minimize the number of critical pyro firings, you still have some *very* critical pyros that, if they misfire, will still result in a Bad Day for MSL.

-the other Doug
tty
QUOTE (dvandorn @ Jul 12 2005, 10:08 PM)
And for those who think that the Skycrane approach enhances the chance of mission success by reducing the number of little pyro actuators that all need to work correctly (as the MERs had to do, in the deployment sequence), that *is* true -- but on MSL, you have to rely on the pyros that separate MSL from its bridle, or else Skycrane will wreck the rover as it tries to fly away.  So, while you can minimize the number of critical pyro firings, you still have some *very* critical pyros that, if they misfire, will still result in a Bad Day for MSL.

-the other Doug
*


Pyro devices in my experience are very reliable if properly stored and handled. However I don't see why "all" would need to work. Since they are also quite light it would be easy to provide redundancy. Admittedly it might be rather too much of a bang if three squibs go off simultaneously, but it would be easy to set them with, say one second delay between firings. If the primary works, then separation has already occurred when # 2 and 3 blow.

tty
BruceMoomaw
To put an end to one red herring I've seen in this thread: the plan is to have the lander drop to 5 meters altitude, hover there, and only THEN reel down the Rover on its cable at 1 meter/second. Lowering it earlier would set up impossible control problems for the throttleable landing rockets. (Even if the Lander screws up and is still slowly dropping as it unreels the Rover, the Rover will still hit the surface with much less force than it would if it was rigidly attached to the bottom of the still-descending Lander with its greater weight.)
Marcel
QUOTE (tty @ Jul 13 2005, 06:06 AM)
Pyro devices in my experience are very reliable if properly stored and handled. However I don't see why "all" would need to work. Since they are also quite light it would be easy to provide redundancy.
*


I don't understand how one can introduce redundancy in pyro's. Pyro's are explosive bolts. You bolt something together with them, connect them to a unit that can make it blow apart by providing a signal. If it does not work, you have a problem. It will still be bolted together. No redundancy possible as far as my knowledge goes...or do i miss something here ? unsure.gif
BruceMoomaw
As for the rate at which the MSL will cross Mars' surface, let me quote the requirements from its 2002 Science Definition Team report:

"The rover shall be capable of traversing up to 3 km in 13 or fewer Sols, using at most one command cycle per Sol.

"The minimum traverse capability of the rover shall be 450 meters per Sol, unless lmited by energy or time constraints. (That is, the degree of onboard autonomy shall not reduce the traverse range below that point.)"

In the "reference mission" cited by the SDT in designing its desired surface operations, the rover travels 69 km over 1000 Sols (with a 33% margin, so the actual time for that traverse is 667 Sols), and makes 23 stops during that expedition for lengthy and detailed study of individual sites.

After this report came out, it became clear that the rover could not achieve all the goals and payload requested by the SDT, and so it was trimmed down further. The actual Announcement of Opportunity for the MSL science payload simply states:

"During periods of traverse, the rover will provide nominal mobility rates of 50 m/sol at a driving speed of 5-10 cm/sec when the vehicle is moving. The total mission traverse capability is expected to be at least 6 km. Target approach algorithms are expected to allow the vehicle, from up to a 20 m distance, to place a contact instrument or tool on a targeted surface feature within 3 sols after identification of the target."
odave
QUOTE (Marcel @ Jul 13 2005, 03:15 AM)
I don't understand how one can introduce redundancy in pyro's.
*


Though I don't know how pyros are put together exactly, I think tty is talking about multiple charges per bolt. If the first charge fails to break the bolt, there will be another charge or two available that could also break the same bolt.
mars_armer
QUOTE (BruceMoomaw @ Jul 13 2005, 06:29 AM)
To put an end to one red herring I've seen in this thread: the plan is to have the lander drop to 5 meters altitude, hover there, and only THEN reel down the Rover on its cable at 1 meter/second.  Lowering it earlier would set up impossible control problems for the throttleable landing rockets.
*


Bruce, I think that was an older Skycrane concept. As of last September, the timeline for the Skycrane maneuver was:

* Chute/backshell jettison at 1000m altitude
* Null horizontal velocity, start vertical descent
* Bridle deployment (~7.5 m length) starts at ~30m altitude
* Deployment complete with rover at ~10m altitude
* Descent stage lowers at constant 0.75 m/sec velocity
* When touchdown is "complete", bridles are cut
tty
QUOTE (odave @ Jul 13 2005, 02:53 PM)
Though I don't know how pyros are put together exactly, I think tty is talking about multiple charges per bolt.  If the first charge fails to break the bolt, there will be another charge or two available that could also break the same bolt.
*


I was thinking more on the way that pyro charges are used e. g. to jettison aircraft canopies. The squib is used to create gas that drives a piston that causes the separation. I. e. the pyro charges and the frangible bolts are physically separate. In this way it is simple to have redundancy though at a certain weight penalty. You could also have multiple bolts each with a charge strong enough to break all bolts, though this might cause asymmetric separation. Multiple charges in a single bolt is certainly possible, though common-cause failure raises it ugly head here.

tty
dvandorn
QUOTE (BruceMoomaw @ Jul 13 2005, 02:21 AM)
The actual Announcement of Opportunity for the MSL science payload simply states:

"During periods of traverse, the rover will provide nominal mobility rates of 50 m/sol at a driving speed of 5-10 cm/sec when the vehicle is moving. The total mission traverse capability is expected to be at least 6 km. Target approach algorithms are expected to allow the vehicle, from up to a 20 m distance, to place a contact instrument or tool on a targeted surface feature within 3 sols after identification of the target."
*

That is actually *less* mobility than has already been demonstrated by the MERs! That doesn't seem right, somehow.

-the other Doug
BruceMoomaw
Well, keep in mind that those official MSL driving requirements were written BEFORE the MERs proved to be so relatively easy to drive. Also, they were an absolute barebones minimum, for scientific planning purposes -- I don't think anyone seriously expects this thing to drive less than 30 km or so across the surface over 2 years.

As for "Mars Armer's" description of the MSL landing procedure: he may be right. I'll have to check with Brian Muirhead (whom I talked to this January about the Skycrane system, but who didn't mention any such procedures). Since there is, I believe, also a Web description of the MSL mission from that meeting of the Mars Roadmap committee, I'll take another look at that too.
dvandorn
I just want to make sure that the re-statement of the driving requirements for MSL didn't result in a re-design of its motors or transmission systems, to *limit* its maximum speeds and ranges to those stated in the final spec.

Of course, if the spec was reduced not by operational considerations but by wieght factors (i.e., if a faster, longer-ranged MSL would weigh so much that you couldn't land the whole thing on Mars), then I can understand it. But if it's a matter of just changing the gearing, for instance, which doesn't cost you a lot in terms of weight, then surely there is enough time between now and MSL fabrication to add some higher-speed options...

-the other Doug
RNeuhaus
I have the curiosity to know about the Skycrane's criterion to select the safest place to land when it is hovering.

Rodolfo
BruceMoomaw
I don't believe it does -- that selectioon is made by the landing-obstacle avoidance system BEFORE the Lander starts reeling the Rover down, since any horizontal movement by the lander afterwards would set the rover swinging, which is exactly what they don't want.
RNeuhaus
QUOTE
since any horizontal movement by the lander afterwards would set the rover swinging, which is exactly what they don't want.

Oppy's landing area
I think that the rover swinging is true. According to the above URL, it shows that the initial path of landing comes from West to East, the prachute was left on the west side, the heatshield was left on the east side due to the velocity effect and its greater weight and the Rover's landing landed on the north west side of the landing path. This means that there were a swing from south east to north west before the pyramide was dropped by the jettisioned rocket from the backlanding shield.

QUOTE
since any horizontal movement by the lander afterwards would set the rover swinging, which is exactly what they don't want.


Then, I suppose that the landing-obstacle avoidance system must have already selected the best area (relativerly free of big stones or dangerous slopes) that has an area big enough to cover all probabilities caused by the rover's air swinging.

The next question is that the tether which will lower Rover have a platform to absorb from vertical velocity.

Will the Rover's main body or legs be temporaly connected to any kind of shock absorbers in order to avoid the use of wheels as direct shock absorber?

I think that the swing landing might be unavoidable and the rover's platform must have a width enough to contrast from the swing velocity and land friction in order to slip without rolling the rover to any side until it stops softly. That additional safeguards might be the best solution for MSL just in the case that the Skycrane might fail to meet the landing requirements. Any additional thoughts unsure.gif

Rodolfo
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2024 Invision Power Services, Inc.