14 ms·
Nothing could have prepared me for how that catch looked. I was sure the rocket was careening into the tower at the last second before it straightened out. The
by ghc 2y ago
Nothing could have prepared me for how that catch looked. I was sure the rocket was careening into the tower at the last second before it straightened out. The control algorithms must be incredible for the landing system to work within those small tolerances.
- starspangled 2y agoThat was one of the greatest things I've ever seen.
- marssaxman 2y agoLiterally brought me to tears, watching that happen.
- ellipticaldoor 2y agoCan you explain why? I don't understand why this is so important
- starspangled 2y agoI certainly can explain it, I'm just not sure I can do it in a way that you could understand. If you yell me what's one of the greatest things you have seen and why, I might be able to better relate.
- monkeydust 2y agoWatched several times, amazing stuff. Had flashbacks to playing Jupiter Landing on the C64! https://en.m.wikipedia.org/wiki/Jupiter_Lander https://en.m.wikipedia.org/wiki/Jupiter_Lander Edit: SpaceX should create a simple 'catch' the rocket game. Play in browser style. Just for kicks and marketing.
- miyuru 2y ago> Edit: SpaceX should create a simple 'catch' the rocket game. Play in browser style. Just for kicks and marketing. there is, it was discussed in the FAA thread. https://news.ycombinator.com/item?id=41821220 https://news.ycombinator.com/item?id=41821220
- renewiltord 2y agoThis is a pretty fun game. Not so easy. Good job to the creator for having added some mechanics that make it work.
- dominicdoty 2y agoShameless plug for my own project with a similar bent - write your own lunar lander autopilot in browser lunar.unnecessarymodification.com
- poorlyknit 2y agoVery cool :) I got a good enough score in the basic scenario by playing around a bit but it would be cool if you could link some kind of tutorial (e.g. to a digital PID video or something like that).
- demarq 2y agovery fun got 51! with fuel limit + mass enabled.
- demarq 2y agoso now managed to save 63% fuel, but score doesn't seem to take this into account (still 51)..
- sobani 2y ago> SpaceX should create a simple 'catch' the rocket game. Play in browser style. Just for kicks and marketing. They did: https://starshipthegame.spacex.com/ https://starshipthegame.spacex.com/
- CodeWriter23 2y agoI succeeded with a High Score of -287!
- KORraN 2y agoThis one is not from SpaceX, but gives way more fun: https://mechazilla.io/ https://mechazilla.io/
- HPsquared 2y agoThat's great, really shows how they are "threading a needle" on the control side of things.
- tails4e 2y agoFelt special witnessing history, was delighted my kids and their friends all were glued to the TV for it also.
- dtquad 2y agoMIMO and nonlinear control theories are probably some of the hardest topics in all of engineering. SpaceX control system also has to compensate for the fuel moving inside their rockets so the control algorithms probably involve some kind of fast numerical fluid simulation. Another interesting thing SpaceX is doing is to use consumer-grade chips in triple redundancy configurations instead of using $100,000+ radiation-hardened aerospace/defense grade chips.
- autonomousErwin 2y agoI always thought the liquid sloshing would be one of the hardest to simulate (considering how chaotic fluid mechanics is). Interestingly, I think this caused the 2nd Falcon launch to fail (the LOX sloshing).
- JorgeGT 2y agoIt is difficult, but there are modeling approaches that work, such as VoF (https://en.wikipedia.org/wiki/Volume_of_fluid_method https://en.wikipedia.org/wiki/Volume_of_fluid_method). Basically, in addition to velocity, pressure, temperature, etc., you store an additional scalar in each cell of your computational mesh representing the liquid's volume fraction. Then, you solve an additional equation to transport that scalar.
- s1291 2y agoSolving the Navier-Stokes equations numerically in 3D is very time-consuming, even on HPC clusters, not to mention the additional modeling required for multiphase flows. Your answer implies that the solutions are obtained almost instantaneously, which is not the case.
- nothercastle 2y agoThey probably pre solved a bunch of scenarios and interpolate between them known solutions
- jatins 2y agotimestamp for "catch" is around 40:00 for those curious
- blackeyeblitzar 2y agoWhen Musk first proposed this, I thought he was crazy. It seemed like something a school boy would draw up. Now I think this will become routine and forgettable after a few more successes. Is there a word for that - something out of fiction becoming mundane?
- api 2y agoI thought it was plausible given the accuracy of F9 landings, though I still wonder how it will work at scale if one failure destroys the landing site. That could ruin the cost benefits. Where his vision hit a lot of speed bumps is second stage reusability. Starship is a beautiful second stage to throw in the ocean. They’ll probably get it landing but the heat tiles will require a lot of refurbishment between flights. They’re going to have to figure something else out.
- edm0nd 2y agoiirc they plan on making multiple launch sites so one getting damaged wont grind operations to a halt.
- DiggyJohnson 2y agoI really don’t think it will ruin the cost benefit as much as you suggest, especially when they have multiple sites and multiple locations. It wound be catastrophic, but they are presumably building Tower like the rest of their hardware, and every time they launch it represents a $100M saving compared to the competition.
- ericcumbee 2y agoIt took months to more than a year to repair shuttle’s heatshield in some cases. SpaceX replaced the entirety of the heatshield with a new design and a new ablative underlayment in a matter of weeks. I suspect they will be able to do it even quicker with design and process improvements. Small scale repairs of the heatshield between flights probably wont be all that big of a deal.
- 2y ago
- drexlspivey 2y agoThe booster was falling at 4500 Km/h 30 seconds before the catch with 2-3% fuel left. How is that amount of fuel remotely enough to stop the downward momentum?
- stetrain 2y agoBecause the rocket weighs a small fraction of what it did at launch, specifically because it only has 2-3% fuel left. Fuel is the vast majority of the vehicle weight at launch, kind of like an empty vs full can of soda.
- justin66 2y agoThe atmosphere does its bit to slow down the booster as well. It’d be interesting to see a plot of the power output over time on reentry but I’ve always assumed the motors aren’t doing a lot of work other than keeping the booster stable, until the very end.
- stetrain 2y agoYes, for most of the booster return it’s ‘gliding’ with the rocket engines completely shut down. They ignite a subset of engines just a few seconds before landing for the final slowdown and maneuvering. Edit: here is a video from further away that shows the rocket gliding in under control of the grid fins before the engines light and execute the final landing maneuver: https://x.com/shaunmmaguire/status/1845444890764644694 https://x.com/shaunmmaguire/status/1845444890764644694
- justin66 2y agoSuper heavy probably makes certain aspects (unrelated to the tower catching the booster) of landing much easier, by virtue of its greater mass. Timing with the Falcon 9 was always essential, because the minimum thrust (with one motor, throttled all the way back) of one of the engines was enough that if you left the motor running, the booster would start to rise. Time the beginning of the final burn incorrectly and you had a real problem, with the booster either crashing or rising at the end and... then crashing, but with Super Heavy, it might simply be a matter of having to compute a slightly different sequence for shutting off the engines. They sure made it look easy...
- Torkel 2y agoI thought the same, screamed out "ouch that doesn't look good!" right before the catch. The last part of the live stream they showed footage from a different angle and there it didn't look too bad though! For sure controlled. Scott Manley put out a tweet that they went down towards a non-tower position until they were at three engine controlled burn, and only then did the side shift.
- brcmthrowaway 2y agoDo the control algorithms use AI?
- arijo 2y agoActually this is a good question - there's a lot of control engineering research on Data-Driven Control Systems. Check Steve Brunton youtube channel, he is one of the leaders in this area: https://youtu.be/gb_C9LcjDSI?si=xUjqUZ9-0MIFohX6 https://youtu.be/gb_C9LcjDSI?si=xUjqUZ9-0MIFohX6
- ThouYS 2y agoIf I had to bet, I would bet against it. Boston Dynamics for example, for the longest time, didn't use anything other than Model-Predictive Control. Only recently have they started using RL
- lutorm 2y agoNo. Way too hard to validate.
- shellfishgene 2y agoThe simulation they show at 21 min into the video is almost exactly like it happend in the end, to it seems it went perfectly as planned.
- pj_mukh 2y agoI know the control algorithms are the mind-blowing part here but, does anyone have any literature about how the Rocket localized itself with respect to the chopstick arms? It must've been some combination of GPS and Radar pings to the arms? And then the onboard IMU to make sure it hits it straight.
- ThouYS 2y agosuper good question, especially with all the tilting involved, which would make visual servoing difficult. Maybe some form of beacons on the ground?
- cdash 2y agoCould just be differential/rtk GPS. You can get incredible precision with that.
- shellfishgene 2y agoFrom the control point of view, isn't this exactly the same as F9 landing on a pad, except the pad is virtual, floating in between the chopsticks and the ground? Or course one difference is that the approach needs to be from the correct direction.
- m4rtink 2y agoI seems to remember some article mentioning the Falcon 9 using radar (+ presumably other sensors) & even having a landing site map uploaded (mainly for the return to launch site scenarios) with prioritized exclusion zones in case of a landing failure.
- schiffern 2y agoHans Koenigsmann alluded to mapping during the CRS-16 post-flight press conference. https://www.youtube.com/watch?v=tVSPCPoc8hs&t=373 https://www.youtube.com/watch?v=tVSPCPoc8hs&t=373
- jessriedel 2y agoA major difference is that F9 (landing on a wide flat pad) had a quite wide acceptable horizontal error, 10 meters or more, whereas I think this (landing between two chopsticks) needs like ~1 meter accuracy in the radial direction.
- stetrain 2y agoHere’s a video from a farther vantage point that gives a better perspective on the landing process: https://x.com/shaunmmaguire/status/1845444890764644694 https://x.com/shaunmmaguire/status/1845444890764644694
- gcanyon 2y agoThat's amazing footage, and you're right about the perspective: from the official feed the distances seem compressed compared to what we see in this footage.
- ralfd 2y agoThe little boy who was lifted on his dad’s shoulders got me emotional.
- Faark 2y agoWow, those pressure waves! Didn't see them yet, on all the close up footage. Thx for posting
- jccooper 2y agoThere's also a very impressive sensor and actuator story.
- slashdave 2y agoYes, indeed. But I will add that the sheer size of the rocket helps in this regard. I think it is rather hard to appreciate the massive scale of the feat by watching videos.
- mrweasel 2y agoA clip from some news program popped up on YouTube, just a two minutes clip of the catch, I was convinced that it was reversed. The fact that this is possible, that they made it work is nothing short of amazing.
- gibolt 2y agoPrior launch landed 0.5cm away from the target location. This is equivalent to landing a 25 story building right on the foundation. Absolutely impressive accuracy and precision.