IM-1 Odysseus lunar lander

[edkyle99]

Is a 7 miles per hour that causes the removal of landing gear "gentle?"

Apparently it was a "brilliant", "highly successful", "soft landing". 

 - Ed Kyle

[theinternetftw]

the spin cycle has commenced.  If they used all the fuel/oxidizer/helium--as just claimed--how did they land as planned under power and not in free fall?

Sounds like they hit the ground and the thrust after the engine bell contacted the ground wasn't enough to make the lander rise back up too much if at all.

There was positive spin in this press conference, but that wasn't it.

[theinternetftw]

Some rough presser notes:

Expected total mission time
  - 144 hours (end this evening)
  - will put to sleep and try to wake upon next lunar day

Used all helium, all methane, all oxygen.

Firsts according to Tim Crain:
 - first time a cryogenic payload has been fueled on the pad
 - first time a flight of an overwrapped linerless cryogenic tank has succeeded
 - first time a methalox engine has been fired in deep space, fired repeatedly in deep space
   - set the record for length of an engine firing with LOI, then broke it the next day with powered descent

Still 'firsts' from Crain? Unclear.
- thermodynamic venting systems for cryogenic cooling
- inline processing of optical imagery, which enabled safe landing
- got multipath data from lunar south pole using multiple radios

Four things they couldn't test before launch
  - fully loaded methalox vehicle through launch loads and vibration
  - free flight control of engine must be tested in space
  - comms network communicating at lunar distance
  - camera performance at the moon

Future
  - more cameras
  - commercial mission called IM-C1
  - larger lander, Nova D
  - comm sats around the moon [Note, these currently planned to fly on IM-2 and IM-3]

350MB of sci/eng data brought down as of now

NASA payloads
  - LN-1 - four passes of comms with DSN
  - ROALSES
    - four antennas to deploy on surface
    - more heat than expected
    - lost one antenna in transit
    - at that point, decided to collect data in transit, so got bonus data
    - also deployed on surface, got data as expected
    - ROALSES team has detected freq of radio noise from Earth; Earth much quieter than expected, but loud at high freq
    - "have all the ROALSES data down"
  - SCALPSS
    - planned to observe plumes during landing
    - had hardware problems that did not allow operation during descent
    - two days ago, IM team worked with SCALPSS team to change operations mode
    - SCALPSS now working
  - NDL
    - all NDL data has been received
    - team reports NDL worked "far better than they expected"
  - RFMG
    - collected data while tanking on the surface
    - also during coast, during descent, and while in lunar orbit
    - first time RFMG has been integrated into prop system
  - LRA
    - retroreflector
    - do believe LRO can range to LRA, will work on that in the coming months
    - now have a location marker on the moon at 80 deg south

  - "Instead of ending up with a few bytes of data, which was the baseline goal for us, we've gotten over 50 MB of data, which, we went basically from a cocktail straw of data coming back, to a boba tea sized straw of data coming back."

  - Says team went from overcoming challenges to "incredible successes for all of our payloads."

The 11 crises
  - the warm methane on the first attempt
  - the star tracker issue
    - too tight of a tolerance check
    - prevented vehicle from getting power positive
  - in commissioning maneuver, discovered a drift in the yaw channel of the main engine control
    - drifted to one side in the CM, saw again in TCM1
    - had to come up with scenarios where the engine geometry would explain that drift
    - knew they could do the small burns, but amount of drift would be too much for powered descent
    - moved the CG estimate around the vehicle around
      - suspected original CG wasn't incorrect, but moving it mimicked the problem signature
    - did patch to redefine the engine geometry
      - under full thrust load, geometry of actuators was slightly different
    - patching it allowed for successful LOI and descent
  - once engine pointing was tuned up, during powered descent, it "looked like a video game" - had never seen it perform that well.
  - engine chill
    - the CM burn, chill wasn't right on oxygen side, had to go again, so didn't fire on time.
    - TCM1 burn, chill wrong on methane side, had to reload, fired a couple hours later.
    - after TCM1, had engine chill handled, next five burns were exactly on time of ignition.
    - if had missed LOI by a minute, wouldn't land on the moon.
  - lower orbit after LOI
    - perilune was low
    - flight dynamics had preloaded ability to do a lunar correction maneuver
    - moved orbit from 100km circular to something close to descent orbit
    - control room moved deorbit insertion from orbit 12 to orbit 3
  - laser rangefinders
    - there's a pin in the cable that you have to insert to activate laser, can't see the pin once inserted
    - clearly fixable
    - vehicle performed, just with optical nav, well enough to land safely
  - most of the crises that were cited were identified and resolved before they impacted the mission

Landing hard
  - it's sounding like coming down hard was a consequence of doing purely optical nav + no good altimeter data + actual landing site having a higher elevation than the notional landing site elevation that was programmed in (I suppose as a last resort).
  - came down with the engine firing, because the Automated Flight Manager hadn't yet switched to the mode where it sensed when to shut the engine off.
  - saw a spike in the combustion chamber, know the engine bell contacted the surface.
  - landing gear took brunt of the load, broke one or possibly two landing gear.
  - sat there upright with the engine firing for a period of time.
  - as it wound down, the lander gently tipped over. simulation suggests it took about two seconds to fall.
  - landed on a 12 degree slope.
  - that 12 degree slope compounded with helium tank or radio shelf results in an angle that's approx 30 degrees off the surface.
  - photo confirms the orientation
  - roll orientation shows that the planned roll to point towards Earth worked
  - top deck solar array shadowed also suggests roll was correct
  - landed about 500m from a 500m diam crater, can see it in the 'lean' photo, use it to see how they're oriented.

Antennas
  - lander pointed in the right direction via roll maneuver, but fell all the way backwards save 30 degrees after leg break.
  - 'and so our antennas are in an off-nominal configuration.'
  - had to really work with lunar telemetry partners, also DSN.
  - absolutely the signals are bouncing off the moon - receiving direct path, plus the opposite polarization from the reflected signal
  - had to sort through how to handle that
  - afterwards could monitor health, send basic commands for 16 hours a day
  - over Australia, could pull down data (from the Parkes 64m dish) for 8 hours at a time
  - at that time, 'no dead air'
  - You want to fix SCALPSS? Great. Be downloading data while you work with SCALPSS.

Success
  - NASA's goal for CLPS was for there to be a soft landing that could gather data. That was done.
  - Science payload goals - still assessing; e.g. SCALPSS didn't gather landing data.
  - SCALPSS and ROALSES will fly on future CLPS deliveries also

Lunar night / end of power for the lunar day
  - will prep lander for lunar night instead of draining entire battery
  - expect 5 hours left before no commanding / no telemetry
  - will leave computers and power system in a place where they can try to wake it up after lunar night
  - main limiter is batteries - that chemistry does not respond well to deep cold
  - 14 days of deep cold
  - battery not tested to that level of cold, neither was the flight computer or the radios
  - "If we asked our vendors to tell us what the probability was of surviving the deep cold of the moon, they would not put it in writing, and well they shouldn't."
  - solar arrays should handle it fine
  - confident arrays will send power, but will batteries receive it and will the electronics be there to use it

Sharing learning
  - "want to give insights of our learning and experience"
  - expects conference papers, talks, briefings about what they've learned, to raise all boats

were *not* transmitting through high gain antenna, using an omni antenna "about the size of a water bottle"
  - could only pull down that data when going from horizon to horizon at Parkes.
  - very affected by wind - if there's wind when moving a 64m dish, it has to stop moving.

Payload success
  - every payload has met some level of their objectives
  - both navigation demonstrators have graduated to actually having been used for navigation - LN-1 during transit, NDL on descent.
  - just a few hours before press conference, got confirmation all payload data has been downloaded
  - will turn sensors back on if lander wakes up for day 2

Laser rangefinder cable
  - it's a range saftey requirement to not have an active laser while getting ready to launch
  - the disable was by removing a pin that then needed to be inserted into a cable
  - a problem with not testing how they fly
    - the engineering units for the rangefinder didn't have the limiter
    - the cable for the testing the flight unit didn't have the missing pin
  - it sounds like maybe by 'pin' they mean a literal conductor pin in the connector interface
    - which would explain how once the connector is connected, you can't see the missing pin anymore

SCALPSS issue
  - a hardware failure in the serial port kept them from collecting data during descent
    - fixed that after landing
  - even if that hadn't happened, due to the flight computer not expecting landing at the altitude it happened, SCALPSS wouldn't have been triggered ("...is my understanding")
  - however, minimum success criteria were met - e.g. surviving launch, doing transit checkouts (did checkouts on surface.)
    - sounds like due to serial port problem, hadn't done transit checkouts actually in transit (or discovered they hadn't)
  - firefly mission has their own version of SCALPSS, will be able to use this data to improve that try.
  - other imagery that was taken during landing IM is happy to share with the SCALPSS folks, have seem some of those, does have plumes.

More on the crises
  - not all had the same 'time constant' to fight against.
  - star tracker problem had a very stark immediacy to it. Only 'a matter of hours' available to solve it before power runs out.
  - Engine start sequence fix (chilldown) had opportunities built into the mission to fix over the coming burns before LOI.
  - Same for fine tuning the engine geometry.
  - If any of them had been left unresolved, they'd become major.
  - But not all of them had such a high pressure time component.
  - "With that said, this is a sprint mission to the moon"
    - fastest transfer since Apollo.
    - so the pressure of only having so many days before you have to be in lunar orbit was definitely felt
    - and so didn't necessarily breathe easier on the ones that had more time; they were critical

ESA
  - was very excited for their mission
  - talking to a number of European companies flying on "our mission 2" (IM-2? May or may not be PRIME-1?)
  - ESA has expressed interest in flying payloads on that mission

Other payloads
  - heard from several companies entertaining sponsorships

More on the rangefinders
  - if they'd had them, they would have "nailed the landing"
  - lander has a terrain map for anticipating terrain elevation, is robust to variations in that terrain, have tested for that
  - "confident that if the rangefinders had been intergrated into the system, we would have absolutely hit the bullseye."

NDL hack
  - succeeded in getting the sensors into the registers the landfinders were going to use
  - succeeded in remapping the new sensor geometry for use
  - however, there was a 'data valid flag' that the original rangefinder hardware had that they had to overwrite to always be valid
    - they didn't overwrite it and so no NDL data was marked as valid for use
  - thus they landed with IMU and optical nav only

Major redesigns
  - the rangefinder fix is not that
  - but does talk about adding cameras and adding antennas

Innovation from IM
  - e.g. their engine injector
  - during development, printed and tested an injector every 10 days
    - printed in 5 days, post processed in machine shop, fire it, characterize it
    - built 40 injectors, worked perfectly in flight

EagleCam
  - successfully ejected post landing
  - camera or wifi not working properly at the moment, Embry Riddle currently working on it
  - would love to fly it again on a future mission

Timelines
  - Didn't start getting telemetry packets from the vehicle until about Day 3 on the surface.
  - That was when they noticed they hadn't processed the NDL data.

Still getting back newly generated payload data as lander continues to survive
  - I assume 'all payload data received' above means anything that was generated and stored

Antennas
  - since they've been on the surface, all comms has been with the low gain antennas

Power
  - in minimum power mode for computer, radios, power distribution units, and heaters, the lander runs about 125W.
  - orientation means the lander dies early, but top deck is pointed toward the east, so should start getting power at sunrise

Temperature
  - as the sun gets lower, things are starting to cool
  - that's giving them information on radio and multipath on the moon in different temperature profiles
  - columbia insolation did so well, they plan on using additional columbia materials on IM-2
    - relationship is now more than a sponsorship, will expand the partnership

NASA payload data
  - after about six months, will end up in PDS

More on radios
  - before they figured out how to config the radios, they were getting "drops of data."
  - then went to, with Parkes in Australia, a "steady trickle of data."
  - when they have access to a high gain antenna (and IM-2 has a better one than IM-1), "that will be a flood."
  - orders of magnitude more data through high gain.
  - plan to eventually have a real time bent pipe link via lunar orbiting assets. At that point can do live video of landing.
    - lunar orbiting assets can also help with safety for Artemis and sharing more day to day info about surface operations.

Landing site names
  - Will have a competition (within IM, they probably mean); Tim Crain threw out Penelope, the wife Odysseus went through his travails to return to.

ESA Lunar Pathfinder
  - comms relay demonstration on a future CLPS mission in partnership with ESA
  - will be dropped off in orbit before landing

[edkyle99]

Revised understanding is that Ody is tilted only about 30 degrees rather than completely on its side.

30 degrees from horizon horizontal, not 30 degrees from vertical, is my understanding.

 - Ed Kyle

[catdlr]

Revised understanding is that Ody is tilted only about 30 degrees rather than completely on its side.

30 degrees from horizon horizontal, not 30 degrees from vertical, is my understanding.

 - Ed Kyle

https://twitter.com/planet4589/status/1762952551572382185

Clarification: 30 deg above the local horizontal, about 18 deg (if I heard correctly) above local surface which is itself a slope.

[abaddon]

Some rough presser notes:

Amazing work; thank you!

[theinternetftw]

Full resolution versions of the released photos.

[sdsds]

This is probably written somewhere on this forum but I was wondering how this lander is keeping the propellant cold? Or is it not that cold and at super high pressure? Is there some information on how they're managing prop temp and pressure?

Presume they're tolerating boil-off having loaded excess LCH4 for the purpose.

So this was wrong.

Tim Crain pretty clearly said Nova-C landers tolerate high pressures in the propellant tanks.

I was wrong yet again. In the Feb 28 telecon Crain referred to thermodynamic venting systems providing cryogenic cooling.

[ulm_atms]

So many are calling it spin and whatnot but seriously...What is everyone's problem?

IM can either be positive about it or negative.  They have been fully transparent IMO and correcting what they said/think as the data keeps coming in.  I don't really understand what more you all want from them here?  Perfect or nothing seems to be the only two things people will accept. 

I for one will always take positivity in the face of problems/issues....always.

Now to see if it can pull a SLIM and wake back up after lunar night.  I know it's not suppose to...but onward positive thinking!

[Eer]

So many are calling it spin and whatnot but seriously...What is everyone's problem?

I've pretty much written off the constant sniping in this thread as an exaggerated "Hrmph!"

If this was a sports statistics column, you'd put an asterisk by the "successful landing, mission objectives (mostly) accomplished" and move on.

Congratulate the team for resilience (and perseverance) above and beyond the call of duty.

[ehb]

Thank you theinternetffw!
I assumed no one dared a blow-by-blow during the presser due to the noise level of this topic.

  - just a few hours before press conference, got confirmation all payload data has been downloaded

That sounds pretty successful to me.
Congratulations IM!

& the tippy wider leg argument (oh i hope i'm not going to multiply the pain) seems moot.
Stronger legs, perhaps vs. they broke and absorbed the energy, but better a working altimeter

[lcs]

They almost fell in.   Terrain maps a necessity, not a luxury.

https://twitter.com/this_is_tckb/status/1762918459376038158

[brussell]

I notice some lack of perspective in this thread. This landing was a WILD SUCCESS! Yes, it was a "soft" landing. Not "soft" would have been a smoking crater on the ground with immediate and terminal end-of-signal. Do you know how hard this is to do? How easy to screw up a little thing and fail completely, like so many other landers or spacecraft have done, with much larger budgets? Some people here are nuts.

I'll be the first to make fun of their hyperbole vs. screw ups, but it was a success. And now they'll fix their problems and will keep putting landers on the moon.

[catdlr]

https://twitter.com/amsatdl/status/1762980249485078833

It looks like we might have seen the last transmission for this Lunar cycle.  There is no detectable signal for longer than the deep fading period now..

[catdlr]

https://twitter.com/stim3on/status/1762966276027117615

I made a panoramic projection of this image, maybe this helps with interpretation.

https://www.360cities.net/image/im-1-landing-wfov-fisheye?override_cache=true

[edkyle99]

They almost fell in.   Terrain maps a necessity, not a luxury.

https://twitter.com/this_is_tckb/status/1762918459376038158

Apollo 15 was on a slope of about 11 degrees. 

 - Ed Kyle

[deltaV]

Revised understanding is that Ody is tilted only about 30 degrees rather than completely on its side.

Why did they need to revise the understanding? Why couldn't they just look at accelerometers to get easy to interpret information on which way is up?

[Remes]

In the presser they said  the landing elipse was 1.5km.

Do we know, what the intuitive machines own lidar looks like? The one which couldn't be switched on?
Looks like it's realy delievering 3d point clouds.
https://ssdl.gatech.edu/sites/default/files/ssdl-files/papers/conferencePapers/AAS%2022-113.pdf

The Nasa Doppler Lidar
https://ntrs.nasa.gov/citations/20220004798
doesn't give you a point cloud which would allow for a reconstruction of the local surface (if I'm not totally mistaken). Actually it is really close to what Apollo used on the LM (one more beam for Apollo, Laser vs Radar but both beeing frequency modulated, and of course nowadays you do your rf in software, not in analog hardware).

But if the intuitive machines Lidar is also switched on on orbit, I believe it would also not deliver a picture of the surface. In the presser they talked about "optical navigation sytem" and "data flow" (1:06:45). I wonder if he meant optical flow? Optical flow allows the reconstruction of the surface. They do use optical flow, see paper below. Albeit that algo would be blind with dust clouds or when the ship is hovering close to standstill. Also it has a lot of complex draw backs.

Found it: https://ssdl.gatech.edu/sites/default/files/ssdl-files/papers/conferencePapers/AAS%2022-113.pdf
It is optical flow. And the outlier rejection is some of the worst and computation intensive problems. And I didn't spend much time thinking about it, but optical flow works on detecting, what image parts have moved to which new location. And sometimes the image parts doesn't move. E.g. in a car there is a point on the horizon you are driving towards, which doesn't move. And while landing vertically the parts directly below you also doesn't move. So quite interesting.

Listening the presser again and started reading the papers: The Lidar would scan the terrain and a terain following algo would deliver the current position. The paper also mentions, that the resolution from Lunar Reconnaissance Orbiter are not sufficient. Still wonder if they could detect the surface based on Lidar only or if optical flow was anyway the way to go.

[Targeteer]

Some rough presser notes:

You win hero of the day for the forum typing that all out. 

[lightleviathan]

(cut down for what's necessary for this post)
Future
  - more cameras
  - commercial mission called IM-C1
  - larger lander, Nova-D
  - comm sats around the moon [Note, these currently planned to fly on IM-2 and IM-3]

Slightly off topic, but noting that the future Nova-D lander, following the Roman numerals trend will have 500kg of payload capacity, similar to Astrobotic's Griffin (~625kg)