SmarterEveryDay SmarterEveryDay

  • 30
  • 911 m.
I explore the world using science. That's pretty much all there is to it. Watch 2 videos. If you learn something AWESOME, please subscribe if you feel like I earned it.

Contact: I currently get lots of correspondence so please forgive me if I'm unable to reply. I mean well, but want to focus on being a better Dad.


36:40Navy SEAL Astronauts - Smarter Every Day 243
Navy SEAL Astronauts - Smarter Every Day 243Áhorf 1,7 m.5 mánuðum síðan
54:33Everything about Sea Turtles  - Smarter Every Day 239
13:30Modeling Murmuration - Smarter Every Day 234
Modeling Murmuration - Smarter Every Day 234Áhorf 890 þ.10 mánuðum síðan
12:56The Gun Detector - Smarter Every Day 225


  • Anyone here getting accession vibes

  • Having stick time is important. That’s the reason we have flight simulator and real stick time instead of 100% reliance on auto pilot and ILS.

  • Awesome video

  • Is it funny how they just "lost" the technology to go to the moon??

  • Yes sir

  • The waiting passive proportionately reduce because tire medicinally concern circa a smiling thistle. shaky, languid toenail

  • They can simulate a lunar lander system now for training, the computer systems and rendering is up to the task That wasn't the case in the 60s.

  • It's a giant nut cracker, Right???

  • So when will this canon be throwing the opening pitch? 😁

  • With the pace SpaceX is developing Starship, pretty soon NASA will just have a test centre on the moon!!

  • The omniscient stick observationally flow because pepper modestly smash qua a magical budget. fantastic, simple doubt

  • backup is always good

  • 16:47 .. that beat tho

  • 16:46 who knew lunar lander test vehicles could beatbox XD

  • Hello 👋 Everyone I am from the future.

  • software with manual override should definitely be the go to. No reason not to make it easier with software, and no reason to not to have an expert ready in case it goes wrong. Redundancy is absolutely worth it when the cost per attempt is so high.

  • landing something on the moon with people in it is maybe the ultimate "trust, but verify" situation

  • “Chug-a chug-a chug-a choo choo!”

  • What about allowing the human to flying as if they are in earth's gravity, and simply having the software translate that to moon gravity controls?

  • I think we need both. Use software, but have human for backup. Software can bug out in ways we have no way to predict where humans can adapt in real time with training in ways software can't yet

  • !t52222111111111111111‰%-+*/45-&&[email protected]&3#$3&$&2$#&43(5:,75*6***44&##$)++=$344&$=((

  • I'm with you Destin. While automated landing systems are nice, we also need to train the people because if the automated system fails at just the wrong time the humans inside NEED to be able to take over. Otherwise the moon may become their resting place.

  • 16:45 That lander is about to drop the sickest new album of 2013

  • 1:43 Among Us?!?!?!?! 😳

  • I want the next sub video !!!

  • That was a sick beat that thing was putting out... I was so waiting for that beat to drop

  • I wonder how that thing could see the ground through the exhaust.

  • This video follows my line of thought right up until you don't figure out the max speed at which a glove can catch a ball.

  • I'll just drop the thumbs up before the ad is even done just in case I forget to hit it after.

  • Destin, new camera in the office? Looks different

  • 14 bis .... Santos Dumont first world !!

  • 90% hydrogen peroxide is pretty dangerous btw... but not as horrible as some other things

  • When Destin's talking about Software vs Human landing, it reminds me of my childhood reading Robert Heinlein scifi novels. I remember many times where a character was hunched over a button or yoke or lever, sweating bullets and years of life as the automated system performed a task, and they had to determine in a split second whether it had botched the job. Man, that took me back!😅

  • I only have one question: how did you get clearance to climb aboard a nuclear submarine as a civilian? Lol.

  • "It was fun :)" - logan 2021

  • Linus: Gives Destin a storage system. Destin: Still uses eHDDs.

    • The drive was from 2013 it seems like. So has just been sitting in whatever long term storage he had at the time. No harm in it sitting there not spinning or being accessed. However, should probably get it on storage that's being backed up so there are multiple copies.

  • If the EGC thruster does not thrust vector how does the lander simulate the angle required to simulate landing on the moon? l love your videos. Thanks for making them.

  • Why didn't Destin put the footage in his NAS that Linus gave him?

  • Hydrogen peroxide is deadly stuff! I was at a cafe near Red Stone with a buddy. Waiter asked what I'd have? I thought it would be cool, so I said, "I'll have H2O!" My buddy laughed and said, "I'll have H2O, too!" I really miss that guy!

  • i want the descent thrusters to be referred to as bangbangs from now on

  • I think it might be like a coupled ILS with the pilot clicking off the autopilot in the last 50 feet

  • Commence dorito consumption

  • Hey Destin, I agree that we should have a trained human pilot, but only as the last resort. If we're looking for redundancy when piloting a lander, it comes down to how many disparate approaches we can apply. I'm assuming they've currently got hardcoded algorithms with some adjustable parameters. That's one option, and the human is another, but I think we should have at least one more in the form of an AI. What we're seeing is that machine learning is an incredible tool for interpreting a complex set of data and deriving the correct result. Unlike the human, the AI doesn't bring any Earth-gravity preconceptions, and it doesn't need to be painstakingly written. Instead, it just tries all the different approaches to evolve towards the solutions. Take it a step further and develop a GAN where it tries to land safely while an opposing AI tries to develop a landscape that is impossible to land on. Then both AI will become incredibly robust in a short amount of time. The opposing AI could also be helpful for human pilots' training. I think that would be the perfect plan B if/when the wrote algorithms fail, or if the pilot is unable to perform. It also means we have a highly adaptive tool that is generalized to conditions we wouldn't expect on just the moon. It could apply to many different missions. I can imagine if we're trying to eventually get lots of people to space, needing lots of highly trained pilots is going to become a limiting factor.

  • As a computer scientist I know how often computers glitch. As a licensed pilot I would not fly a craft that doesn't let me take control. I will never ride in an autonomous vehicle unless there is some way to take over the controls.

  • Are these government sponsored videos?

  • 1:28 "Thats what my buddy Logan did . . . He works at NASA" he says that so nonchalantly

  • John 3:16-17 16 For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. 17 For God sent not his Son into the world to condemn the world; but that the world through him might be saved. Romans 6:23 23 For the wages of sin is death; but the gift of God is eternal life through Jesus Christ our Lord. Romans 10:9-10 9 That if thou shalt confess with thy mouth the Lord Jesus, and shalt believe in thine heart that God hath raised him from the dead, thou shalt be saved.10 For with the heart man believeth unto righteousness; and with the mouth confession is made unto salvation. Amen!

  • Rick rolling Nassa's applications

  • I think for any new technology like a lunar lander, it is critically important to build replicas that humans can control for training. If we ever get to a point where we go to the moon enough that we gain a full understanding of it, then build full simulators that will suffice. But for now I think building replicas are super important.

  • And didn't explode.

  • I think it would be way cooler and the pilots involved would all say we should land with humans, however, as SpaceX and others are showing, software is quite reliable now when you have good hardware around it, so it’s likely to be safer when you remove an extra set of human error variables. That said, it may be worth it to have a human land it, regardless, just to say we did it... I mean, a human would still be responsible for writing the code and all of that and therefor for landing it, it’s just way cooler to have a human at the sticks, and may be way more meaningful. As long as we get there, get it done, and get it done safely though, it may not matter exactly how we did it in the long run. That said, obviously there maybe should be a human pilot override if the software goes haywire... rather than just aborting at that point which would likely be the case if the lander is automated. Obviously, there are still software mistakes because humans program them, but you can iron out those things with software automated test flights, but things like Boeing having set their clocks wrong on their docking capsule show that maybe there should in fact be a human override, unless we think aborting once we get to the moon is an acceptable outcome, rather than having another shot if the software glitches out or crashes or has a programming error... which unfortunately I think may be the likely scenario in the unlikely event that we do have a software malfunction on landing phase.

  • Human in the loop is ideal. Electronics fail.

  • A good example of how the future of space will be accelerated forward by people like Elon Musk and SpaceX. They would have done this test in a couple of months in the SpaceX back yard like Boca Chica without 100 people and a nuclear style countdown and moved on to the next problem. NASA probably spent years and years and many tens of millions of dollars, plus, as was pointed out, there were multiple overlapping multi-million dollar NASA projects doing the same research. Sadly, today, NASA by itself couldn't move fast enough to get to the moon again before 2050!

  • Thank you for yet another amazing video ! Maybe it could be a middle ground, like new dji fpv drone. Pilot could abort landing spot choice by an algorithm and just point at where she wants to land. Then another algo just executes

  • Love that you got the Browns stadium in there, Cleveland Ohio!!

  • 16:46 - if DaftPunk gets ahold of that you'll get a million in royalties!

  • I think lessons learned from flight school should apply for this landing. Planes have auto-pilots but they do not eliminate the human element even with these advanced systems. There is an important reason for that Like with one of the mars landers. A difference in freedom units and matric units caused it to crash. If a human was on board they could of said this doesn't feel right I'm aborting or taking control. With out joy-stick training the humans will be doomed if 1 calculation is off by a degree. With so much riding on this, not having that back-up for your back-up could be disastrous.

  • @ 6:46 But there comes a point where they break away also. It might be when they get too close to each other. One bird will break off and a group of others will follow.

  • You must have run through the various thrusters role about five times ... makes for a longer video i suppose

  • Hey Destin! Unrelated question but are you still actively researching the characteristics of a cracking whip? The question popped into my head when you showed us that sneak peek at the end of the video.

  • Look at a video working on flight control. I finally recognized someone on the video and I didn't realize that Mike H. works in controls (he coached my son in soccer). Being a Chemical/Control Engineer, I think the control loops from sensor to thruster is a fascinating subject.

  • That's what AI is good for.

  • You may only speak while holding the little plastic lunar lander module. Those are the rules.

  • LLTV idea, a giant quadcopter with a pilot module in the middle. The module would be gimbaled to accurately rotate to the equivalent angle experienced on the moon.

  • I can't believe you've had this footage for years. It must've been so tempting to upload it in the past

  • Haven't we pretty much let go of human controls/intervention for launches, though? I feel like landing is probably eventually going to be machine only too (though be it next 10 years or 100 who knows)

  • Hoping to see Astrobotic feature in this lunar landing series!

  • we going to the moon again !!!!!!!!!!!!!!!!!!!!

  • Ofc I think they should implement both flight control software and human training. However, I’m not confident that humans would be better at solving flight control problems faster than a computer or back up computer could. With proper training I think being able to take manual control is great, however, I’m still very unconvinced that there wouldn’t be even more problems created by manual control. It really depends on what the actual lunar module would be like.

  • Thanks, Destin, for bringing us the kind of science that leaves me in awe, and makes me proud, at the same time.

  • Better beat than anything that has won a grammy in the last 4 years at 16:47

  • Nope. We don't need a stinking emotional chaos-ape overriding the automated symptoms. If we have an emergency that the system can't handle, fine, for now, let the human steer, but that's not an emergency, it's a failed design. We need better and better (and always more better) automated systems to override / anticipate / prevent every problem, including human mistakes. Robotic cars are still in their infancy and they are orders of magnitude safer than humans. AND they don't get tired, distracted, angry, inpatient, or drunk. Space travel is infinitely harder than ground travel, and it needs every advantage.

  • The beard/mustache looks great.

  • like a tesla there should still be some at the wheel/ controls that knows how to operate the vehicle in case something goes wrong.

  • What happens with the gravity cancellation thruster if the vehicle pitches? Wouldn’t it be no longer aligned with gravity? Wouldn’t that create a problem?

    • Good question. Unless it could be gimbaled, which I don't think they talked about.

  • 16:45 This new Front Line Assembly track is fire.

  • come to X

  • Man oh man, this is cool! Man oh man I'm enjoying this video! I'm actually understanding what the heck they're saying, and man oh man, I think this is super cool! This is me (a natural PottyMouth) being very respectful to Destin and the audience😅, but, MAN OH MAN, I'm loving this! After watching Perseverance's landing, this is just what I needed to keep that hype going in my soul! Almost like that School Pride assembly in high school after you win the Division Championship.😄

  • 16:47 is the intro to some kind of dubstep/EDM track

  • But can it run FalcoX ;)