Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Crew Dragon Docks with ISS

spacenews.com

Post reply on HN
Print view View original

181–190 of 260 posts

Re: Crew Dragon Docks with ISS

#181

Post by bkanber »

landa wrote: post #85
moftz wrote: post #50

Earlier quoted context omitted.

Very little in terms of what you would consider AI or ML. It's a lot of trig and control theory. You do a lot of the math beforehand to figure out the script the craft will follow to reach the ISS. Once it gets close, you either use manual controls or let autopilot do the docking. The autopilot is basically looking some dots on a camera and firing the thrusters to make the dots line up right. It's more like a dumb PI…

What you wrote is what I would have guessed . Are you also guessing or do you know for sure? It sounds right, but I'm very curious about the actual answer to this question.

I don't work in the industry, but I have the training, and can confirm what he said. No AI or ML. We use a combination of closed-form solutions and deterministic numerical analyses. There is no actual need for AI or ML in this problem domain, outside of perhaps the design phase. Sometimes dynamic programming and some ML techniques are used when, I dunno, figuring out the best configuration for an antenna or profile of a nozzle. But flight dynamics and control do not need AI, as the physics are well-understood.

Re: Crew Dragon Docks with ISS

#182

Post by pfdietz »

kbenson wrote: post #171
pfdietz wrote: post #114

Earlier quoted context omitted.

Guidance systems are vastly less expensive now, and can take advantage of things like MEMs, fiber laser gyros, and GPS (and, of course, modern computing power). The tanks on Falcon are welded using Friction Stir Welding, which was invented in the early 1990s. The landing algorithm for the first stage uses convex optimization algorithms based on interior point methods, which were not available in the 1960s. However, s…

> Simply evolving the Saturn 1B (which was cheaper, per lb of payload to orbit, than the Saturn V) could have easily beaten the Shuttle's economic performance. I don't know how much it applies, but sometimes evolving something requires switching out components for other ones, which are more expensive. Not everything scales. Marketing isn't the only reason why 50% more capability often costs 100% more. I imagine with…

We have an example of a better component, that is otherwise very similar: the first stage engines of the Saturn IB and the Falcon 9:

H-1: https://en.wikpedia.org/wiki/Rocketdyne_H-1

Merlin: https://en.wikipedia.org/wiki/SpaceX_Merlin

The Merlin has considerably higher specific impulse and thrust/weight ratios. It also has very fast throttling, a necessity for vertical landing. And, it's remarkably cheap (less than $1M per engine). I don't have the figure for the inflation-adjusted cost of the H-1, but I suspect it was much more expensive.

Re: Crew Dragon Docks with ISS

#183

Post by _0w8t »

the_duke wrote: post #70
toyg wrote: post #53

Earlier quoted context omitted.

I love space exploration like the next guy, but I have a massive fear that a proliferation of space-stations will lead to unpleasant situations. With all its tensions, the ISS has been an amazing tool for soft diplomacy; if we start having country-specific stations we’ll soon end up with nasty arguments. Already we have competing satellites rumoured to have classified capabilities, I hate to think what they could do…

China plans to launch the first part of their own space station next year. [1] They actually wanted to join the ISS effort, but were given a firm no and had no other option than to do pursue their own. [2] I don't think we will see any other country-specific stations for a long time, but modular commercial stations that can be shared for research and potential business endeavors (hotels, specialized manufacturing, ..…

I read Russia considered to disconnect their module from ISS when the agreement about ISS expires.

Re: Crew Dragon Docks with ISS

#184

Post by ozim »

ericol wrote: post #9
flixic wrote: post #4

You can try docking yourself at https://iss-sim.spacex.com

Took me 3 tries to do it; my bad was not carefully reading the instructions.

Am I kind of weird? Did it first time without reading anything just looking at numbers on the HUD. Maybe a bit of common sense?

Re: Crew Dragon Docks with ISS

#185

Post by acheron9383 »

7thaccount wrote: post #22
bane wrote: post #15

SpaceX runs basically as a giant R&D program that also happens to accomplish useful things as byproducts of its approach (with the miracle of insurance to cover costs if things get explody). Every flight is data for incremental improvement. If you think of all SpaceX has ever done was try to get to this point, with somewhere around 80 flights on the Falcon 9, but for a variety of mission buyers (NASA, USAF, Commercia…

Incredibly impressive, but you have to take into account both more modern technology, as well as the fact that they were able to build off the shoulders of giants. It is still amazing of course, but I doubt they could've done the same thing during the era of the mercury program, so not really apples to apples comparison.

Yeah. It is great that Nasa and the US military paved the way, like has happened before (The internet, etc). SpaceX feels like a new era of spaceflight, harnessing so much of the potential that US government created in the previous era.

Re: Crew Dragon Docks with ISS

#186

Post by usrusr »

barbegal wrote: post #168
pfdietz wrote: post #114

Earlier quoted context omitted.

Guidance systems are vastly less expensive now, and can take advantage of things like MEMs, fiber laser gyros, and GPS (and, of course, modern computing power). The tanks on Falcon are welded using Friction Stir Welding, which was invented in the early 1990s. The landing algorithm for the first stage uses convex optimization algorithms based on interior point methods, which were not available in the 1960s. However, s…

> The landing algorithm for the first stage uses convex optimization algorithms based on interior point methods, which were not available in the 1960s. Is this level of optimisation really needed? I thought most of the landing was having enough fuel margin and then running a PID control algorithm during the landing burn.

I believe that the biggest obstacle in SpaceX's "hover slam" landing isn't calculating and controlling the burn but getting sufficiently precise input in time for those calculations. It must be a miracle of sensor fusion.

Re: Crew Dragon Docks with ISS

#187

Post by fma »

hinkley wrote: post #36

I recall Elon was hoping to get an order of magnitude reduction in cost per launch, but Wikipedia claims it costs $160 million per launch versus the Soyuz price of $76 million per astronaut. That would “only” be a 3x reduction with a crew of seven, but they’ve configured for a crew of four, which is a hair under 2x. I wonder how reuse affects the math, and what they’ll be able to do to lower those prices further. Obv…

Per this interview in 2012:

“There were times when I thought he was off his rocker,” Mueller confesses. “When I first met him, he said, ‘How much do you think we can get the cost of an engine down, compared to what you were predicting they’d cost at TRW?’ I said, ‘Oh, probably a factor of three.’ He said, ‘We need a factor of 10.’ I thought, ‘That’s kind of crazy.’ But in the end, we’re closer to his number!

https://www.airspacemag.com/space/is-spacex-changing-the-roc...

This is in response to the engine. Maybe they got the engine price 10x cheaper, but the other components not so much?

Re: Crew Dragon Docks with ISS

#188

Post by acheron9383 »

paulryanrogers wrote: post #101
trixie_ wrote: post #90

Earlier quoted context omitted.

Maybe not during Mercury, but definitely during the 70s all the technology was there to build a cheap, reusable rocket and capsule. The dream was that the shuttle could be that system, but it did not turn out that way. And giving in to the 'sunk cost fallacy' they supported the shuttle for the next 40 years. Arguably you could say we are 50 years behind where we should be. Unfortunately driving down costs is probably…

Didn't Space X also benefit from NASA launch facilities? I imagine their payroll also includes quite a few NASA trained people too. They certainly aren't doing rocket science from scratch or in a vacuum. Not too say this isn't a big deal. Just that it's not all Elon and his wunderkind.

Yeah of course, SpaceX would be stupid to not use the resources NASA will make available for them. Honestly, NASA would probably much rather be out of the business of building rockets, so they can focus on the next generation of space experiments. The less money NASA has to spend on hauling stuff into space, the more they can spend on what they put up there.

Re: Crew Dragon Docks with ISS

#189

Post by jeffdavis »

TeMPOraL wrote: post #133
toshk wrote: post #119

Earlier quoted context omitted.

At the same time "startups" are also not able to reach any of the tangible results musk is producing. He seems to be able to benefit from the resources of a big company, while at the same time keep the fresh mindset of a relatively small company.

He has a goal he wants to accomplish with this company, and unlike almost anyone else, his goal isn't to enrich himself and spend it on luxuries. He really wants people to colonize space. An actual non-monetary goal is what's missing, IMO, for most companies.

It's about having any specific goal, really.

Money is more of a general goal, like happiness. Hard to stay focused.

Re: Crew Dragon Docks with ISS

#190

Post by pfdietz »

barbegal wrote: post #168
pfdietz wrote: post #114

Earlier quoted context omitted.

Guidance systems are vastly less expensive now, and can take advantage of things like MEMs, fiber laser gyros, and GPS (and, of course, modern computing power). The tanks on Falcon are welded using Friction Stir Welding, which was invented in the early 1990s. The landing algorithm for the first stage uses convex optimization algorithms based on interior point methods, which were not available in the 1960s. However, s…

> The landing algorithm for the first stage uses convex optimization algorithms based on interior point methods, which were not available in the 1960s. Is this level of optimisation really needed? I thought most of the landing was having enough fuel margin and then running a PID control algorithm during the landing burn.

Apparently yes. See Lars Blackmore's chapter in the 2016 "Frontiers of Engineering" (National Academy Press) (free registration required)

https://www.nap.edu/read/23659/chapter/10

https://www.nap.edu/download/23659

"SpaceX uses CVXGEN (Mattingley and Boyd 2012) to generate customized flight code, which enables very high-speed onboard convex optimization."

This is actually a great example of a true NASA spinoff, btw. Lars Blackmore was at JPL before moving to SpaceX. Landing on Mars and landing a first stage back on Earth aren't the same problem, but they're close enough for the work to transfer.

Post reply on HN