Working Yondu s Arrow That Flies When You Whistle - Change Direction In Flight

This video is sponsored by KiwiCo. So this is Yondu's arrow from Guardians of the Galaxy. If you haven't seen it, here's what it does. Basically an arrow that can change direction based on Yondu's different whistles. And you guys want me to build it in real life? Like you really want me to build it in real life. Thousands of requests for this one. And I've honestly been trying to put it off for a bit just cause it doesn't seem physically possible. So if I'm gonna spend all this time working on something that might not actually end up being possible, all I ask is that you leave a project idea down in the description for something I should build next. Like now that I'm building this, I need another idea that thousands of you want to see me make. Also thumbs up ones that you think are cool so I know it's popular. And of course, you know, like and subscribe if you want. All right, so let me just read you the wiki article that explains how this works. A quiver of arrows composed of yaka, a special sound sensitive metal found only on Centauri 6. The yaka arrow can actually change its trajectory in response to certain high octave whistle sounds some Centaurians can produce. Yondu is so skillful at controlling his arrows, he can cause the arrows to return to his hand or weave through a crowd of people without touching them. Sound sensitive metal. We are so screwed. What does that even mean? Let's reel it back in a bit to the realm of what is physically possible. I think the whistling control part is doable. Like you can definitely parse out frequencies of a whistle using some physics. Write a program that turns those sounds into a control. But the hard part is gonna be making an arrow fly. For that, we need a way to provide lift to the arrow. Planes have wings, rockets have rockets. The reason it's gonna be so hard is there's no space for any of this on an arrow. Like the back fletching is the only part that actually sticks out at all. So I'm sure you all know this, but the fletching on the back of the arrow is what makes it fly straight. Like if it starts to get off target, wind puts higher pressure on that side, correcting it and bring it back straight. Super simple, self-correcting, it's been done for thousands of years. So if we wanna turn an arrow, what if we just turn that fletching? Like we get a servo, a transmitter, a receiver, 3D print some parts. Boom, RC arrow with controllable fletching. This could work. We're just doing some control tests to show my bow skills, make it look like kinda cool, you know? We'll do some of this. We have the control, I hit like six feet off the target. Now your job is gonna be to steer this to make it hit the target. So go left, nice. So you just kinda wanna watch as I'm shooting this and try and. Oh yeah. First try, dude, we're probably gonna be done with this video after this. Try not to steer it into yourself either. All right, we're gonna try and shoot a little longer of a distance so that everyone might have time to actually change direction. So we're gonna shoot at those trees. Try and just turn it that way. All right, put some weight on the front. Hopefully this will make it less tail heavy. Oh, I did. Really? Yeah. Turned? Yeah. Doesn't have enough wind to actually make the turn. Does turn in the air, I think, you're right. It's not looking too good, but I did get some comments saying to use rockets to make this. So, you know, yeah, I'll try your suggestion. Your request, put a rocket on here. Do you think it's gonna work? No. Flat out, no. It landed back end in somehow. So that means it did turn. So this might work if we made the arrow super light. Like we might be able to eventually get it to turn slightly if we shot it far enough. But to make it really work, like we need some thrust vectoring rockets or some SpaceX. And even that wouldn't get the same control as the movie. So I think rockets are out, unfortunately. What we really need is more lift. Like normal arrows only stay in the air for a couple of seconds max. And Yondu's arrows look like they can stay in the air as long as you want. Let's try wings. Problem with just adding wings is now it's just an RC plane. So we need to add wings without, you know, adding wings. And I think the fletching is gonna be our best bet. Like we could probably scale it up a little bit without it looking too silly. And as I said before, normally it's used for just keeping the arrows straight. But I think if we do it right, we should be able to get a little bit of lift out of it. Also, I should mention I'm super new to RC planes and stuff. I don't know why you guys asked me to make this when you got Peter Striepel or flight test. But I'll give it my best shot. And I think I'm gonna work backwards for this one. Like I'll start with a model airplane, get that flying, and then slowly try and turn it into an arrow. So I started with just some ordinary paper airplanes because that's what I know best. Folded a few and eventually got one that flew pretty well. Then to motorize it, I took apart one of my mini drones I actually had left over from my last video, flipped the propellers from vertical to horizontal, put them on the wings right along the center of gravity. That center of gravity though, that's kind of important. Like that'll be a theme for this video. So after some failed attempts, I eventually got the plane to stay in the air for a little bit. But I mean, it already wasn't great and I hadn't even tried to turn it into an arrow yet. So I thought maybe some foam would be a better material to use. So again, I just started with a classic plane design, put the motors on. It actually worked really well, like surprise, surprise. But yeah, now we just gotta stretch out the design until it turns into an arrow. So I printed out a picture of an SR-71 Blackbird because that's the plane that looks most like an arrow in my opinion. So I cut it out and used a foam cutter to put it on foam. Guys, guys, if you don't have a foam cutter, get a foam cutter. It is so satisfying. Anyways, I put the RC motors on it and did a little test flight. It kind of worked actually. I mean, it wasn't perfect, but you know, there was definitely some potential. So I stretched out the design even further and cut it all out. This is where we start to have to think about weight placement because we want the center of gravity over the wings because that's where the lift's coming from. Problem is all the arrows weight is in the front and little tiny fletching, you know, wings are in the back. So I made the nose foam real thin and tried to keep the wings and back a little thicker. Also moved the battery to the very back. And before I put the motors and everything on, it actually was flying not that bad, but as soon as I added that extra weight, I don't think the wings would generate quite enough lift to keep it in the air. So I don't think wings are looking too hot either in terms of lift capabilities for this arrow. Rockets were too unstable, so I guess that leaves propellers. Specifically drones. You know, I've actually got a bit of practice building drones. I've built tiny drones all the way up to human flying drones. So maybe there's a sweet spot, you know, somewhere in the middle we could make like an arrow shaped drone that actually works. So I hopped on SolidWorks and sketched up a rough model that might work. So I was thinking, if we use enough of these tiny little drone motors, we should actually be able to fit them along the body of the drone while still keeping the arrow like relatively thin. I did some rough calculations and figured out that each one of these motors can produce at least a couple grams of thrust. So if we use 11 of these motors, we should be able to get around like 30 grams of thrust. So if we can make a drone that weighs less than that, should be good. On paper that is. So I printed out this model. Gonna put motors in each one of these little holes, but you know, I was just looking at it and then realized this is horrible. This is way too complicated. You know how much math would need to be done to get this to work? Like, this is a very weird drone configuration. Doesn't even look that cool either. It looks kind of like a... Yeah, moving on. So at this point I've dipped my feet into a bunch of different pools. I do think there is slight potential for each design, but they would all take a bunch more work. Plane would work if the wings could just give a little bit more lift. Drone would probably work if there wasn't so many complicated parts. So what if we combine the two designs and replace the wings with tiny little drone motors? Like intuitively at first, it doesn't seem like it'd work. Like if we just stick a giant arrow on the front of these drones, it's probably gonna flip over, right? Well, it all goes back to that weight distribution I was talking about. Like, you know those statues or those toys that balance, but they shouldn't look like they balance, but they do because of the weight placement? Well, we can use that here to our advantage. We just have to design an arrow that has the center of gravity all the way in the back where the drone is. Then it's just a matter of keeping the total weight low enough so that the drone can actually still lift it. So I ordered a tiny mini racing drone. I think it's just over three inches in size. So it's still pretty small. Most arrows have fletching that's about like an inch or two in length. So three inches, gonna be on the larger side, but you know, not out of the realm of possibility. So next step, we need to test the lifting capabilities of the drone. So I first started off with a binder clip, obviously can lift that. So I used it to clip a 10 gram weight to the bottom of the drone. No problem. Let's go for 20. All right, not bad. So I whipped up a quick prototype with my foam cutter, and this actually only weighs three grams. If this works, we got a lot of room to play with. Oh, no freaking way! What? Are you kidding me? What? How does that work that well? Whoa! Tension right there. We just gotta make it a little beefier. Yes, we are definitely onto something. So I ordered up some lightweight balsa wood and made another stronger, better prototype. So this thing actually worked like very well. I was pretty surprised. It doesn't really look like it should fly. The only thing I'm not super happy about is the fact that it's made of just wooden foam, you know? So it can't really fly through anything or do any real damage. So I did a little bit more digging and pulled the trigger on a real racing drone. Brushless motors. And this thing is way more powerful. It actually looks a little smaller, too. So I did a weight test with this, and it's able to pull almost 150 grams. So we can definitely make a rigid arrow in that weight limit. Couple 3D prints later, and we have ourselves some arrows that can be screwed directly onto the drone. So I basically printed the back end way more dense than the front. I also put the battery back there as well, along with some more counterweights. So now the center of gravity is right over the propellers. Also tried to model it slightly after Yondu's arrow with the sloping front and back end. It was a little bit on the wide side, but you know, if I saw this, I would still definitely think it's an arrow. So last thing we gotta do, make it fly when I whistle. We'll get to that in a second. For now, let's just take this out and see if it actually works. But real quick though, before we fly this, I wanna take a quick second to show off some of the projects that you guys built and sent to me. Check them out. Always love to see what you're making, guys. I'm a huge advocate for getting creative and building. I really do try and spread that message. I think we need a lot more of that in this world. So if you've made anything cool, definitely send it my way, at Jayler's video on Instagram, or you can email me too. Also, to help you get creative and get building, definitely check out KiwiCo, the sponsor of this video. Their message is right alongside mine, inspiring young inventors. So they've got tons of DIY project crates that include everything you need right in the box. They actually have a lot of crates that have the same stuff as in my videos. This wire art is the exact same stuff I used on my drone. Their colored fire kits, I've used the same stuff in multiple videos. Like the way I started building was through stuff just like this. Like it's a great way to do some learning that's actually still fun at home, especially in these times. They'll send you crates each month, and if you check out the link in the description, you'll get 50% off your first month for any crate. Just check out KiwiCo.com slash Jayler's video. You know what the holiday's coming up? Great gift idea. So big shout out to KiwiCo, great company, great message. Check them out. Definitely support what they're doing. All right, now let's fly this thing. So there we go. We know it flies pretty well, actually. So just to put the cherry on the cake, let's make it actually whistle controlled. To do this, I'm gonna take the pitch of our whistle, use it to make the drone fly forward or backward. I think anything more than that, like up, down, left, right, would just really overcomplicate things, and I don't think it'd be that easy to do. So I started by writing a Python program that displays the frequencies of our whistles. And when I say I wrote a program, I mean steal pieces from other people's code and put them all together. I'm kidding, I'll link all my sources down below. I'm not a programmer. But I do know a little bit of physics. So all whistles are made up of a bunch of different frequencies. So to separate out each frequency into simple waves that we can measure, we can use a physics operation known as Fourier's transform. Basically, it takes any function of time, so like us whistling for a while, or like any sound for that matter, and separates it into its constituent frequencies. Yeah, I do physics, whatever. When I say I do physics, I mean I downloaded a plugin that does all the calculations for me. You know, it's still something. Now if we do a little whistle, you see the top graph is just the raw sound. The bottom chart is the Fourier transform. And the numbers right here are actually the parsed out data. So you can see the graph changes based on the pitch I do. Whistle. Now what we gotta do is pass the signal data to an Arduino, make it move a servo, to move the joystick on the controller to tell the drone to move forward. Now yes, I could have used Arduino to send the signal directly to the remote, you know, not using the servo, send it to the drone, sure. But I still wanna be able to override this and just grab the knob and move it, because otherwise we might crash the drone. But yeah, let's give it a shot. So high note. Nice. All right. So I mean, I think that's it. I think this arrow now does pretty much everything the movie does. So now just for fun, let's try to recreate that scene where it like flies around and flies through everyone. So we set up some targets and let her rip. I think I got an idea. Yes, that is what we need. We need you, specifically. What, what part? You know, you know what's going on. I just need you, not for, I don't, I don't wanna say the reason I need you, but you're the perfect person to use. That's so sweet, all right, cool. Yeah, yeah, let's go. No, you're gonna love it. You're gonna love it. It's gonna be really sick. All right. Yeah, yeah. Yeah, it's kind of like a car. This is what we need. This is not exactly, like I thought if you're gonna do something for me, it'd be for my personality. Oh, you wish. Oh, yay. Blue man group, you need a fourth guy? All right, why don't you go ahead and rub that in. Oh my God, is it satisfying you? Yeah, it's like a spa. So there it is, guys. Let me know how I did. Did take a lot of work though, so if you wouldn't mind hitting that thumbs up button, really would appreciate it. And yeah, leave your ideas too. As I said before, stuff you wanna see me make next, drop it down below. Once again, shout out to KiwiCo for sponsoring this video. Wow, that's a nice flow. All right, thank you guys for watching and we'll catch you in the next one. Peace.