The more interesting question is why it crashed, or rather:
- why did it eject its parachute too soon?
- why did the brake thrusters activate for a shorter time than planned?
Does anyone here know if the landing steps were timing-driven or telemetry-driven? Was it supposed to eject its parachute after N seconds, or when reaching M meters above the surface?
The radar was only activated after the front shield was ejected, so there was no altitude data prior to that event. It could be that accelerometers on board sensed the dynamic pressure, so that parachute deployment was not purely open-loop. I'll try to dig through publications to find out what the EDL (Entry-Descent-Landing) strategy was.
Thanks! So it looks like the entire descent was telemetry-based (source [1], section 3.4):
The functions of the EDM GNC [Guidance, Navigation and Control system] are (1) to detect the
conditions adequate for the deployment of the
parachute and (2) to command and control the final
powered descent phase. The GNC sensors aboard the
EDM are the Inertial Measurement Unit (IMU) and the
Radar Doppler Altimeter (RDA). A main challenge for
the EDM GNC relates to the fact that the EDM needs
to go into a hibernation mode during the 3-day long
coast phase. During this phase, the inertial reference of
the IMU is lost and an additional sensor is necessary in
order to rebuild the inertial attitude shortly before
entry.
Some of it may come from a different engineering culture. There is an argument that the Russians have better launch vehicles and engines because they were willing to take a more rapid prototype approach and blow up a relatively large number of engines to get the kinks worked out. The US took a more conservative approach with fewer but more highly engineered engines that ended up being less efficient than some Russian designs.
The problem is that you can't do rapid prototyping on Mars landers, each iteration is just too expensive. This is where the US approach works better.
As for the ESA, I have no idea. Hopefully they got enough data out of the lander before it crashed to figure out what went wrong exactly.
I'm not sure I agree with that argument. At least comparing specific impulses of Proton and Soyuz vehicle engines with (older, pre-Russian engined) Delta and Atlas shows the latter are often significantly higher.
That's not to say the Russians haven't done good engines but RD180 for example is derived from the Energia engines which didn't have (at least publicly) lots of trials/failures.
Well, SpaceX is taking the rapid prototyping/incremental improvement idea to a new level, and intends to land on Mars. Their first attempt or three are likely to fail. But would you care to bet on whether SpaceX or the US government will have successfully landed more spacecraft on Mars within 30 years?
Some of those failures happened during the faster, better, cheaper era at NASA, in which taking on more risk was not a deal-breaker. The pendulum has swung back. For more context, see: http://www.thespacereview.com/article/35/1
well the GP said "landing", so that disqualifies your mention of the Mars Climate Orbiter and the Mars Observer, and Deep Space 2 was carried by the Mars Polar Lander, so that really only counts as one loss.
The statement was not "it's actually weird how US has only had success at landing on mars". It's still difficult to land things on another planet. The expense and the distance limit quick iteration. So...pretty successful, yeah.
I am pretty sure that they simulated a lot this descent. This failure means they do not have (yet ) the correct modell.
Or the descent sequence has to many variables and any slight change might propagate down the computation to a larger deviation ( like the exact time to stop the thrusters - seems they stopped earlier than needed ).
Or we need to load more advanced algorithms/AI software that decides and computes a local solution. At this moment, we don't really know.
Schiaparelli was a disk with a crumple designed to land on Mars. The disk interrupted its signal one minute before landing and scientists are trying to explain why: descense too fast, incorrect model, unexpected failure, anything else.
Euros sent a probe/lander to Mars, it crashed. The orbiter it traveled with, which has it's own mission to analyze trace gasses in Mars outer atmosphere(I think) is working fine.
"In the decades since humans started sending spacecraft to Mars, the Red Planet has outscored us handily; humans have only about a .411 batting average overall -- not great for missions that cost billions of dollars and countless time to build. Seventeen landers have been sent, and just seven made it to the surface safely, each of them with varying degrees of success."
Afterwards, Curiosity rover luckily succeeded, under much harder constraints then these now (being huge but sensitive), so that is at least 8 from 18. With this one, 8 from 19 (giving, interestingly, 0.42). I don't know which the real ratio now is, am I missing something?
Reminds me of a classic The West Wing scene, when a missile test misses its target by 137 miles. Yes, every step is important, but 6/7 is still not a success.
LEO
That nine out of ten criterion that the DOD lays down for success in these
tests were met.
BARTLET
The tenth being?
LEO
They missed the target.
BARTLET
[with sarcasm] Damn!
LEO
Sir!
BARTLET
So close.
LEO
Mr. President.
BARTLET
That tenth one! See, if there were just nine...
I think it's almost a tongue in cheek thing in aerospace. There's stuff like "RUD" which stands for "Rapid Unscheduled Disassembly" and means the rocket exploded, etc.
In that same spirit, one could even say there was just a slight deviation from the expected lithobreaking landing phase - so all steps were indeed performed, just not to spec
Schiaparelli was never a science mission. It was a technology demonstrator to see whether the landing system worked. Even if it had landed intact, it had a tiny set of science instruments and would only have lasted a few hours.
Well, we now know that there are a few bugs --- but we also know that most of the landing system works, including hibernation, terminal guidance, reentry, heatshield deployment, the parachute... all of which is hugely valuable data. Knowing that the last few stages didn't work is even more valuable, particularly as it's returned telemetry on what happened.
Focusing only on the fact that the last couple of steps failed and it crashed is doing the mission a huge disservice. Schiaparelli did precisely the job it was designed for.
Schiaparelli, the ExoMars Entry, descent and landing Demonstrator Module (EDM) will provide Europe with the technology for landing on the surface of Mars with a controlled landing orientation and touchdown velocity. The design of Schiaparelli maximises the use of technologies already in development within the ExoMars programme. These technologies include: special material for thermal protection, a parachute system, a radar Doppler altimeter system, and a final braking system controlled by liquid propulsion.
Schiaparelli is expected to survive on the surface of Mars for a short time by using the excess energy capacity of its batteries.
Come on. All the people who worked on those features designed them with the intention that they would work. All of them would have been happier, and better regarded by the rest of the world, if everything had worked first time.
It's just the way life is. Not every setback is a disaster, but it's so much better if things work smoothly. Nothing succeeds like success.
It's important in science to not sugar coat scientific conclusions as you might sugar coat PR communications (ie. how can we present this in the best light to get funding for the next trip)
The objective was to demonstrate an ability to land in a useful condition so the next probe could land with experiments using this technology. Everything else is an in-process metric. The probe crashed and didn't land.
31 comments
[ 3.1 ms ] story [ 67.7 ms ] threadSchiaparelli has crashed
- why did it eject its parachute too soon?
- why did the brake thrusters activate for a shorter time than planned?
Does anyone here know if the landing steps were timing-driven or telemetry-driven? Was it supposed to eject its parachute after N seconds, or when reaching M meters above the surface?
EDIT: Here are some papers worth persuing:
[1] https://solarsystem.nasa.gov/docs/Bayle_ExoMars_EDM_Overview...
[2] http://ippw2016.jhuapl.edu/docs/abstracts/Missions-abstracts...
[3] http://arc.aiaa.org/doi/abs/10.2514/6.2011-6341 (behind journal paywall)
The functions of the EDM GNC [Guidance, Navigation and Control system] are (1) to detect the conditions adequate for the deployment of the parachute and (2) to command and control the final powered descent phase. The GNC sensors aboard the EDM are the Inertial Measurement Unit (IMU) and the Radar Doppler Altimeter (RDA). A main challenge for the EDM GNC relates to the fact that the EDM needs to go into a hibernation mode during the 3-day long coast phase. During this phase, the inertial reference of the IMU is lost and an additional sensor is necessary in order to rebuild the inertial attitude shortly before entry.
It's another strike, it's actually weird how much success the US had at landing on mars. The Russians have constantly failed at doing so.
Meanwhile in this universe : https://en.wikipedia.org/wiki/Mars_Climate_Orbiter https://en.wikipedia.org/wiki/Mars_Polar_Lander https://en.wikipedia.org/wiki/Deep_Space_2 https://en.wikipedia.org/wiki/Mars_Observer
The problem is that you can't do rapid prototyping on Mars landers, each iteration is just too expensive. This is where the US approach works better.
As for the ESA, I have no idea. Hopefully they got enough data out of the lander before it crashed to figure out what went wrong exactly.
That's not to say the Russians haven't done good engines but RD180 for example is derived from the Energia engines which didn't have (at least publicly) lots of trials/failures.
(Elon plans to be living there long before then.)
well the GP said "landing", so that disqualifies your mention of the Mars Climate Orbiter and the Mars Observer, and Deep Space 2 was carried by the Mars Polar Lander, so that really only counts as one loss.
The statement was not "it's actually weird how US has only had success at landing on mars". It's still difficult to land things on another planet. The expense and the distance limit quick iteration. So...pretty successful, yeah.
Or the descent sequence has to many variables and any slight change might propagate down the computation to a larger deviation ( like the exact time to stop the thrusters - seems they stopped earlier than needed ).
Or we need to load more advanced algorithms/AI software that decides and computes a local solution. At this moment, we don't really know.
Earth is 25-29 against Mars.
Sorry, I was in a hurry and shortened Europeans to Euros. Probably could have said ESA.
http://www.popsci.com/science/article/2012-08/mars-rover-cur...
"In the decades since humans started sending spacecraft to Mars, the Red Planet has outscored us handily; humans have only about a .411 batting average overall -- not great for missions that cost billions of dollars and countless time to build. Seventeen landers have been sent, and just seven made it to the surface safely, each of them with varying degrees of success."
Afterwards, Curiosity rover luckily succeeded, under much harder constraints then these now (being huge but sensitive), so that is at least 8 from 18. With this one, 8 from 19 (giving, interestingly, 0.42). I don't know which the real ratio now is, am I missing something?
Completing "most steps" in a landing, but not the last one (I.e. Safe touchdown) is equal to a crash.
LEO That nine out of ten criterion that the DOD lays down for success in these tests were met.
BARTLET The tenth being?
LEO They missed the target.
BARTLET [with sarcasm] Damn!
LEO Sir!
BARTLET So close.
LEO Mr. President.
BARTLET That tenth one! See, if there were just nine...
Schiaparelli was never a science mission. It was a technology demonstrator to see whether the landing system worked. Even if it had landed intact, it had a tiny set of science instruments and would only have lasted a few hours.
Well, we now know that there are a few bugs --- but we also know that most of the landing system works, including hibernation, terminal guidance, reentry, heatshield deployment, the parachute... all of which is hugely valuable data. Knowing that the last few stages didn't work is even more valuable, particularly as it's returned telemetry on what happened.
Focusing only on the fact that the last couple of steps failed and it crashed is doing the mission a huge disservice. Schiaparelli did precisely the job it was designed for.
Schiaparelli is expected to survive on the surface of Mars for a short time by using the excess energy capacity of its batteries.
http://exploration.esa.int/mars/46124-mission-overview/
It's just the way life is. Not every setback is a disaster, but it's so much better if things work smoothly. Nothing succeeds like success.
The objective was to demonstrate an ability to land in a useful condition so the next probe could land with experiments using this technology. Everything else is an in-process metric. The probe crashed and didn't land.