Quick thinking and a stroke of luck averted a Moon lander disaster | Hacker News Reader

Quick thinking and a stroke of luck averted a Moon lander disaster

techcrunch.com

19 points_justinfunk19 comments

> The navigational issues started after Odysseus conducted a planned maneuver called lunar orbital insertion on Wednesday night, which put it in an elliptical orbit around the moon. That ended up being extremely “fortuitous,” Altemus said, because it led mission controllers to try to use a navigational subsystem called “laser rangefinders” far earlier than planned (the lasers were going to be activated for the first time during the final descent phase).

> After reviewing the data, the company realized the morning of landing that the lasers were not working — because they did not turn off a physical safety switch on the component while it was still on the ground.

And then they happened to have the NASA LIDAR package they could use as back up. And then the lander tipped over, but not on the side with the important equipment on it. Good lord they were lucky.

And EagleCam was not deployed at landing due to complications of the LIDAR software change, so instead of releasing EagleCam and not seeing the landing (due to the unexpected lateral movement), they can use the camera to confirm the status of the lander. So lucky.

so they cross strapped a payload laser sensor to use for attitude/ranging control because the laser range finders didn't remove its hardware interlock?

I’m amazed they had performance data of the LIDAR package useful for calibrating into attitude and range finding processing? And they updated the software in a few hours to use that new data into the navigation models? That is crackerjack software programming; I would guess it’s a higher level language onboard than ada to allow such rapid prototyping and deployment?

I bet that they had it already developed and validated and just needed to enable it, either by flicking a config or updating to a different build with the other lidar enabled.

1/20/24, Japan’s lunar lander settled upside down. 2/22/24, the US lunar lander settled sideways.

Why can’t we make it with a shape that is foolproof, like maybe a sphere?

Or better yet, a gömböc.

Ha! Today I learned

At this rate, someone should launch a small rover up there that just tips everyone’s landers upright

There's a sci-fi short story about the NASA (?) bureaucracy sending an "auxiliary maintenance system" (re: astronaut) to fix its Mars lander robots when they break, even though the same agency refused to send a person. I can't remember the author; knowing 90s me, it'd be Asimov, Clarke, Pournelle, Niven, etc.

It'll be mass. Self righting mechanisms are a thing and probably quite easy in low gravity, but they have mass, and every kilo you spend on making the machine better is one you haven't spent on carrying stuff up. Very like the aerospace problem where making aircraft heavier is worse, except that you're climbing out of the gravity well by throwing stuff out the back of the rocket which makes the ratios much worse.

What about just distributing existing mass better? I used to have a little oblong toy that was unbalanced like this. No matter where or how you threw it, it would tumble over and eventually reorient straight up due to the weight all on the bottom.

weeble-wobble. https://en.wikipedia.org/wiki/Weeble

The optics on the machine falling over are horrendous. The CEO lying the model down on the side for a large stretch of the press release yesterday interacted badly with the stock price.

I've just about convinced myself that the falling over is interesting but unimportant. There was a statement that the panel under the vehicle contained "passive experiments" which didn't need to leave the machine. That probably means the various experiments it's supposed to run can happen anyway.

The fixed antennae being pointed in an unhelpful direction seems bad for bandwidth but that could mean send the experiment results back with higher priority than photos of the surface which also sounds alright.

I'd really like to know whether NASA are happy with the run (i.e. paying for it and keen for the next) or putting on a brave face for the press. Interested to see what this community thinks about it, reddit is having signal to noise challenges.

Obviously the optics are horrendous, it's why they had to go to the backup optics :)

I thought it was weird that they didn't do any testing of this primary system until they needed it to correct their orbit. One would think dry running all of their components as soon as feasible in the mission plan would be SOP.

Further discussion on the mission:

Private company landing on the moon today

https://news.ycombinator.com/item?id=39465230

Intuitive Machines successfully lands on the Moon

https://news.ycombinator.com/item?id=39474909

Autonomous lander design should consider self-righting capabilities, otherwise "I've fallen, and I can't get up!"

if horizontal velocity of Odysseus had been absolutly zero I guess it would have all gone ok. So was it use of LIDAR laser guidance & hacked software that allowed 2m/sec horizontal vel. while touchdown? A lot we [ I ] do not know.

it is bloody obvious that they should be using self righting landers at this stage.

I don't think they have watched enough episodes of Battlebots to learn that lesson as of yet.

Reply on news.ycombinator.com