 Here, we live in the future. Today, you can go to Maplin in Milton Keynes and you can buy a quadrotor. Then you can bring it back here and you can fly it in that field over there. I say that, I suspect by the end of the day they might be sold out. This is the reality we live in. Quadrotors have become a thing. I see a few dotted around here. I see several flying around the TMF. If you want to go out and buy when you can, as I already said. But what are the other options? So you can go online and you can buy bits and pieces, you can buy kits, you can put them together and you can be flying either within an hour, within a day, within a week, within a month, within a year, whichever route you choose. So a few years ago it wasn't quite like that. You could go out and buy all the bits, you couldn't really go into Maplin and buy a quadrota. But you could go out and buy bits and pieces. You could get an Arduino and you could download an early version of AduPilot and tinker with it until it works and then fly it. But that's not what I did. I decided that I wanted the supreme challenge of trying to do it, or maybe not all of it, but as much of it as I could myself. And that is what this talk is about. Hopefully by the end of the talk you will all want to go away and build your own quadrota from scratch rather than buying one from Maplin. So maybe there will be some left after all. But even if you're not that way inclined, I do want to give a bit of education about how a quadrota works so that even if you don't build it yourself, you will have some appreciation of what they actually do, how they work. So, that's a quadrota. It's one I made earlier. Which bits of it could you make yourself? Well, fairly obviously if you can see the picture. Actually can I just check? Can everybody see the picture okay? I know we've had some issues with this in this tent. I shall have to ask the technical people here if they can do that, but I suspect the answer is not yet. Good, good. So, watch the streamed version of the talk later on and you will be able to see all the slides. So, obviously I built my own frame. Now, lots of people ask how did you choose the size? That's what I made it out of. This I bought it from B&Q. It's an aluminium box section. You probably can't see down the end of it because it's too small from there. And it comes in one metre lengths. So, that is how big my quadrota is. Apparently these work better if you keep them. Right, I took some aluminium sheet and bolted it together. So, that's about all I'm going to say about building the frame. Because that bit, while it was interesting for me to learn a little bit of basic metalworking, that is basically all there was to it. I'll come on to why in a bit. Other things you could make yourself. There's the electronics, the flight controller and all the other bits and pieces. There's the motors. Now, I don't actually recommend trying to build your own motors because modern motors are really, really powerful and really, really light. They are so good. And I think even if you're really good at motor winding, you will really struggle to achieve the performance of a motor you can buy. Similarly, propellers, they're very lightweight, they're very cheap, they're very strong. And they're consumable. So, a few people have tried 3D printing propellers, but they're not that great compared to the ones you can buy. So, personally I wouldn't waste the effort in those. The motor controllers, on the other hand, I didn't make my own, but there is a significant challenge there. And I know that that is something that some people have tried to do. And the other bit is the radio. I had plans to do something partially my own for the radio, but I decided that the rest of it was enough work. So, the radio is one that I bought off the shelf. Okay, but there's still quite a lot of stuff in there. So, what can we do with that? Yes, why a quadrota is kind of a question that some people ask. So, why did I make a quadrota? Well, let's talk about a few different types of aircraft. So, everyone's familiar with an aeroplane. Probably most people in here have been in one at some point in time. And yes, you can get ready-controlled ones as well. Fairly simple physics. You have a propeller providing thrust. You have wings that convert some of this thrust into lift. That keeps it off the ground. And the one remaining problem is control. This is something that the early pioneers of aviation didn't all get right. Some of them managed to get the aircraft off the ground and then realized they didn't know how to control it. So, you need control. And on aeroplanes, control is done by having some moving surfaces, ailerons, elevators, flaps, rudders, and so on, that deflect the wind and change the amount of lift on each wing or on the front versus the back of the plane, and so on. So, the other types of aircraft are rotorcraft, of which the obvious one is the helicopter. Now, the helicopter doesn't have the same control surfaces. It has one big rotor to provide lift and a few other things. It also has a tail rotor, if you have one big rotor above that provides lift, well, there's also some torque coming from that. And if it's going one way, the helicopter's going to be spinning around the other way. So, you have a tail rotor or some equivalent arrangement at the back of the helicopter, which counters that effect and allows you to fly straight. So, what about the equivalent of the control surfaces on an aeroplane? How do you control a helicopter? Well, this is where it gets a little bit complicated. So, I'll try and keep it simple for you. As the rotor blade moves around, under a rotor blade here, right, certain parts of it, the angle of attack, so that is the angle at which the rotor hits the air, changes. So, you can have a high angle of attack on this side and a low angle of attack on the other side. So, you get a difference in how much lift you get on each side, or front and back, or anywhere around this. Now, this is all done mechanically. And as a person who is not that good at fiddly little mechanical bits, that wasn't so attractive to me. So, I didn't want to build a helicopter. Now, I'm not that great with small little mechanical bits, but I am good with electronics and firmware, which brings me to a quadrotor. So, a quadrotor, now it's got more propellers, it's got more rotors, so that's more complicated, right? No, the advantage of a quadrotor is that you can get away with a really simple fixed-pitch propeller as your rotor. So, no moving parts except for it spinning around with a motor. So, that's great for me because I don't think I'm going to like making all those tiny little mechanical bits, but I can control the speed of four motors because I can do that with a microcontroller. And that's all you get to control with a quadrotor. So, how do you actually get that to do things, to do the things you want? So, you've got four rotors here. Now, with a helicopter you have this problem of having to balance it with the tail rotor because otherwise the helicopter will start spinning. Same deal with a quadrotor, which is why two of the rotors have to go one way and the other two have to go the other way round for balance. Now, here's where I use a prop. If we've got a quadrotor here and we have two propellers spinning faster, let's say the two nearest to me, and two propellers spinning slower, let's say the two nearest to you, this is what's going to happen. That was slow motion. What's actually going to happen is this. Okay, similarly, if these two rotors start going faster and these two start going slower, you get that. To make it go up, you just have to have a total amount of lift that's more than its weight. And to make it go down, you just reduce the lift. So that is the simple theory so we can all go home now. Yeah. Yeah, there's a rider on that. Unfortunately, you don't get away with that quite so simply. You get some imbalances, you get some instability. Now, I don't know how clear this picture is. I don't think it's really clear enough for the people at the front seeing the projector. I don't know how that's come across on the TV. What this is showing is a test that I did before I'd written the flight controller just to show what happens. Now, this thing, I don't have this one anymore. I was flying this at EMW last year and it went a bit wrong and it ended up in the water. So this is its replacement. But this is based on the same basic specs, same motors, same props. What this was, before I'd written the flight controller I wanted to test the motors and the props to make sure that it worked properly. So I strapped it down to the heaviest object I could find. In this case, a SparkStation One Plus. In the corner here is one of the feet of the SparkStation One Plus. Now, in this picture, I've got three of the rotors going at full blast and one of them in the far corner is off. So it's unbalanced. And I was not expecting this. But it started lifting the SparkStation One Plus off the floor. Not entirely off the floor. One bit of it was still on the floor in the opposite corner. But that was enough to scare me. I didn't dare repeat the experiment with... Well, no, I did repeat the experiment with only two running, but I didn't dare hold a camera at the same time. So if we're going to write a control system because you couldn't just wire this up to a radio even with the mixing to get the two to work together, you would not be able to fly that. Even a pretty good pilot would not be able to fly that. Because unlike an aeroplane, it doesn't really have inherent stability. That thing is trying to fall to the ground and you are trying to stop it falling to the ground is basically how it flies. So how are you going to control it? Well, you're going to control it with a radio control transmitter or something similar. Now, what about it are you going to control? There are a number of common flight modes that various different flight controllers implement. The easiest to implement is what's called acro or acrobatic mode. That's what people are flying when you see them do all sorts of silly maneuvers out there, flips and all the rest of it. Now, there are actually some modes that make flips easier, which are combo of these two, but these three. But the basic ones are here. Acrobatic mode lets you control the rate of rotation. So if you've got your stick in the middle, it's not rotating. It can be there, it can be there, it can be there, but it's not rotating. If you've got your stick slightly to one side, it's going to be maintaining a constant rate, which is quite hard to do like this, because that's the rate of rotation. In reality, it's going to be more like that, but that's hard to demonstrate. And the same goes this way round or this way round. So that's acrobatic mode. That's quite easy because all you need to do is measure the rate of rotation and I'll come on how to do that shortly. The next mode is called stabilised mode, which is a little bit more useful because acrobatic mode is quite hard to fly in, to fly with. But it takes a lot of effort to learn it and you will crash while you're trying to learn that. So stabilised mode controls not the rate of rotation, but the attitude of the aircraft. The attitude is which way it's facing, which orientation it's in. So this is, as near as I can get it, a level attitude. And that's what you get in stabilised mode if you let go of the sticks. That's not going to stop it from drifting along this way with the wind, or back again, or any of the different variations. So it's going to move around. You're going to have to correct its position, but at least it will correct the attitude for you. In practice, you still get, in this axis, in the yw, you still get the same controlling the rate of rotation. And the other mode, if you've got a really fancy one with GPS and everything else, they call it loiter mode, is where it controls the position for you. So all you've got to do is point the stick forward and it will just go forwards, or point the stick backwards, it will go backwards. As much as it needs to, to compensate for the wind and everything else and just go to where you want it to go. So that's the easiest mode to fly, but it depends on GPS and various other things. So even if you've got that mode, you always want to learn how to fly in stabilised mode at least. So stabilised mode is what I implemented on this. I may one day do acrobatic mode if I'm feeling brave enough, but I don't really fancy a weekend in the workshop rebuilding the frame and strapping everything back onto it too often. So there we go. So, for stabilised mode, if you're going to control the attitude, you need to know the attitude. So how do we do that? Well, this is probably about the right time to look at the, an overview of the electronics that you get on a quadrata. So you have some sensors on the left-hand side. You have a gyro, a gyroscope. This is what's called a rate gyro, as distinct from a traditional spinning ball of metal. And I'll come on to what that does in a minute. And you have an accelerometer. A gyro and accelerometer are the bare minimum you need to fly in stabilised mode. In acrobatic mode, the bare minimum is the gyro, which is why I said it's easier to implement. And I've left a box at the bottom to represent all the other sensors that you might add at a later date. Then you have a control system, and we'll go into the details of what a control system does later, because this is, I think, one of the more interesting bits of the quadrata. And you have the radio. You're using all your commands from the pilot, which turns out to be quite important. From the control system, you send digital signals to ESCs, electronic speed controllers, or motor drivers. Those convert a digital signal from a microcontroller to pulses with lots of volts and amps that go into your motors and make them turn round. And then you have the motors. Four of them. So that's reasonably straightforward as an architecture, I think, except you've just got lots of ESCs and lots of motors. So what can you do with that? If we squash the ESCs and motors together and the sensors together, you end up with this. A control system, a control loop. You have sensors. That's your gyro, your accelerometer, and anything else that you put into it. And I've split the control system into two. The control system feeds the actuators, i.e. the motors. But in the middle we have estimation and we have control. Now this is a very important distinction because the control system will control some variable, in this case, attitude, to be what you want it from the radio. But if you don't know your attitude, that's not going to get you anything. So we'll talk a little bit about the estimation process first. Now the first thing you need for estimation is to figure out a way of representing your attitude. Now this is the maths bit of the talk. There's not much of it. And once this is over, we'll get back on to big boxes. And you don't need to fully understand all this to get the gist. So we start off with a coordinate frame. Some way of saying, you know, which way is north, which way is east, which way is down. This is the standard coordinate frame used for aviation, used for aircraft, but you don't have to use that one. So in this diagram, I put the north being this way into the screen, east towards me, and down being down. So the diagram on the left is some base orientation that you can choose. But typically align that with the ground. That's quite a good way to start. On the right, we have a different orientation, a different attitude. In this case, I've chosen to rotate it 30 degrees just to make the numbers line up. And we can represent the transformation between those in a number of different ways. A few of those ways suck. And a few of those ways work well. The ones that suck suffer from a problem known as gimbal lock, where you can get into certain orientations where you're restricted in which way you can go next. So don't do that because if you end up trying to do some aerobatics or something like that and you end up in one of those orientations, you will then crash. The two most popular ones are this, which is a rotation matrix, also known as a direction cosine matrix. And the other one is quaternions, which I'm not going to talk about now, but they are also equally good and have different advantages. So this is a 3D rotation matrix. This represents the rotation from the coordinate frame on the left, i.e. the ground, to the coordinate frame on the right, i.e. your aircraft. I'll show this another way. So that's the one on the left and that's the one on the right. I've rotated it 30 degrees in that direction. No, I haven't. I lied that way. Right. So this rotation matrix has a few interesting properties, which turn out to be quite useful. If you look at these rows here, the top row shows the coordinates. So these lines are unit vectors. They are of length 1 in whatever system you define. So the rows of the matrix show you the coordinates of the ends of these vectors in that coordinate system. So it's a good way of reading off which way your axes point in the other coordinate system. And the same vertically, they show you the coordinates of the left-hand coordinate frame in terms of the right-hand coordinate frame. Now, it's misaligned in the transition to windows. Oh well. So this turned out to be quite useful because you've got a nice vector that tells you which way is down or which way it thinks is down. In this case, that's showing you this vector in terms of the coordinate system here. Now, that's great because if you've got a vector that tells you which way gravity is pointing, that's a vector in this coordinate system that's pointing straight down. So that's this column here. And if they're different, then you know you've got some correcting to do. I think that's the end of the maths. Right. So let's talk about the sensors. A gyro, or a rate gyro, measures the rate of rotation. It will tell you how many degrees per second you are rotating in each of the three axes. If you're flying in acrobatic mode, that's all you need. You can say, I'm rotating 10 degrees a second. I wanted to be going 12 degrees a second. I need to go faster. Job done. Or at least you can pass that into your control system. If you want to get attitude, it's a bit more complicated. So it tells you your rate of rotation. And if you read for that sensor, say 100 times every second, then every time you read from it, you can add your reading onto your matrix. By add, I mean apply a rotation. So you're rotating your estimate of where you think you are every time you get a reading from the gyro. That's really accurate over a short period of time. So every time you've rotated, you know how much you've rotated, you've kept track of it. Now you know which way up you are or which way is down. Except if you wait over a long time, that's not going to work so well because every time you get a reading, you have a small error in that reading. And those errors add up. And after a while, you think that you're this way up when actually you're that way up. That's not conducive to, say, flight. I actually had this last year. I had a bug in my flight controller where it wasn't doing the corrections to this. And I was actually quite impressed that it could take off and it could hover for a while. And over the course of about a minute, it was slowly drifting to about there, which was the limit of what I could compensate for with the controller. So my flights were limited to about a minute. But I was surprised it was that long. So yeah, there's no absolute reference. There's nothing to correct it with. There's nothing to tell you which way is actually down compared to which way you think is down. So that's where your next sense has come in. Your accelerometer. Now your accelerometer measures acceleration, right? Except gravity is kind of the same thing. So in effect, you actually get a combination of acceleration and gravity. Now, I said acceleration plus gravity. It's actually kind of acceleration minus gravity because if you're not moving, that's equivalent to accelerating upwards by 1g. Right. We make an assumption here. This is about the only assumption we really make. No, it's not the only assumption. It's the most important assumption we make. And we assume that overall your average acceleration is zero over the long term. You might be flying around and everything, but you're not constantly accelerating. Otherwise you'll end up in space. The only time that's violated is when you're in a continuous, long, sweeping, coordinated turn, which quadrotor pilots really don't do very often. And if you do that, your quadrotor will hate you. Now, the downside of the accelerometer, because you could say, well, why can't we just, you know, take the accelerometer reading and we know which way is down? Because that's what a lot of phones do in other applications and so on. Very noisy. Very noisy indeed. You know, it could be all over the place and that's not going to end well for your flight. So what you actually need to do is take a combination. I'm not going to go into the details of these because that is quite mathsy, but you take one of a number of algorithms to combine the gyro and accelerometer. I put some names up so that when you look at the slides later, you know what to Google for. So back to our control system. We've done the estimation. We know what the attitude is. Now it's time to control it. So the most common form of control used in quadrotors today is a PID controller. It's not the only one by any means and it's not even the best, but it is simple. It looks like there's a bit to it, but there really isn't. It's called PID because there are three parts to it, proportional, integral and derivative. And all that's saying is you take your estimate if where you are and you take the signal from the radio which says you want to be there. You've got about 30 degrees of error. You multiply that by your P term, by your P coefficient and that says, okay, I want to apply quite a lot of force to go back the other way. Well, if you're only a little bit off, it just gives you a very small amount of force. That's never going to get you quite there because it's going to slow down as it gets there. So you have this I term where you add up the error. The longer you are pointing this way, the more of a kick it's going to give to bring it back. And the D term just takes your very short term, your very sudden gust of wind and everything else and just locks it in, okay? So how do you tune those values? Because you've got to come up with those P, I and D values. I'm not going to go into the details of that because there's a million things on YouTube on how to do it. I do have a very short video which shows... You try to stack it down to something that's not going to be there. This is the first time I've done something that's not going to be there. You can see that absolutely not on YouTube. I eventually ended up doing it by picking the thing up and holding it and trying to cut myself, but I'm not really recommending that method. How do you debug it? Because if you're writing the control software, you're going to have bugs in it. There is no question about that. You are going to have... All the sensor data, all the radio input, all the motor output, the attitude estimate, every single time it goes around the control loop. That's great because I can look at it afterwards and then figure out what went wrong after it's crashed. Some people choose to do this by telemetry sending that data down the radio. You have to compress it a bit. You have to be more picky about which data you actually send, but that's another way to do it. The other good debugging technique is using your eyes. Look at it while it flies. Is it wobbling? Is it oscillating? Is it stable? Is it sluggish? When it goes wrong, watch it as it drops to the ground and crashes, even though that will break your heart. Once you've got all that data, you can try doing some simulation as well. This is where I'll show you my next video. That really doesn't show up too well. This is real data from a flight at OM 2013. Last year, last summer in the Netherlands. I had a bug in the attitude estimate. On the left here, we have a quadrata that is displaying the attitude from the log file of which way up it thought it was. On the right-hand side, there's another quadrata displaying the estimate as done by a new test algorithm on my PC. I didn't even need to fly it to get that and see after a while they diverge. They're starting to diverge, but you really can't see it, so we'll move on. You're getting close to the point where you can actually fly your new quad that you've just been building. There's a few safety concerns which we really have to cover before we can go out flying. The first one is obviously choose a safe place to fly. We have a flying field here. Unfortunately, the wind's a bit problematic, so that makes that difficult, but you really want to be in a place that if you have to suddenly turn it off or anything goes wrong, it can crash to the ground without killing anybody or maiming them, because that would be bad. The other thing that might happen, the first time you turn it on, it might be confused. The first time I turned this on, for real, in somebody's garden, was... There was a... Actually, it wasn't this one, it was the last one. It was a couple of days before EMF last time. There was a bug in it. It thought that it was rotating like this. I turned it on when it was about here. Or when it thought it was about here. So what's the first thing it did? It tried to correct itself. Now, what did that look like in practice? Well, it was kind of more like that. I was quick. I killed the throttle and I grabbed it when it was up at this angle. But that's... Yeah, I was quick. That could have been quite uncomfortable. This thing is actively trying to kill you. If you have it down on the ground and it is powered up, it could turn on all its rotors and fly into your face. You do not want that to happen. So, yes, it's worse for... Yeah, it's worse for you than that if the quadrador does that. So, when you power it up, you want to do a bunch of pre-flight checks in your software. This thing will not power up if it thinks it's rotating more than a certain speed. This thing will not power up if it thinks that it is more than a certain number of degrees from the horizontal. This thing will not power up under a number of different circumstances. If it doesn't have a radio signal, this thing will not power up if the throttle is not set to zero first. All of these things mean that when I plug it in, I have some degree of confidence that it is not going to try to kill me or anybody else. That's a really good thing to have in there. I would not be wanting to fly this if it didn't have that safety feature. Radio failsafe. A lot of radios support some kind of failsafe mode. If you lose contact, if you go out of radio range, what do you want it to do? Either it has GPS and it can find its way home. Mine doesn't yet. If you don't have that, you need to do something safe. Honestly, the safest thing is probably just to switch off all the motors, let it fall from the sky because you chose a safe place to fly, that means there weren't any people underneath it. The alternative is to attempt to keep flying without knowing where you are and it might go over a bunch of people or vehicles or property or something else and if it then crashes into the ground, propeller is still running or possibly into a person, you really don't want that. So a combination of a safe place to fly and being able to kill the motors at any time. If you lose radio contact or also manually, if you've lost control of it, just kill the throttle or press the kill switch depending on which way you've set it up. The other bit of safety, quadrotors and many other model things use lithium polymer batteries. They are awesome, but if you treat them badly, they will treat you badly. If you overcharge them, if you over discharge them, they don't like it. The consequence, you can look up on YouTube but it involves high pressure gas coming out and catching fire. So that's great. If we are operating in the UK, we just have a little bit more stuff to consider. There is law about this. All aviation in the UK is regulated by the air navigation order. I'm not going to go too much into details of the law because it is quite a complex subject, but I'm going to try and simplify it for you as much as I can just to have a nice safe area to keep it safe. I'm not a lawyer, this is not legal advice. Look it up, etc. The Civil Aviation Authority has published a couple of documents that are relevant as CAP 393, which is their interpretation of the air navigation order. The one specifically relevant to us is CAP 722. I've put these on here so that you know what to Google for later. CAP 722 refers to unmandarial systems, which these are. Of course, in other countries, the law may vary. So the easy way to stay legal is to keep within these guidelines. Keep your aircraft less than 7kg because then you don't have to comply with lots of extra regulations. 7kg is quite a lot. This weighs just under 2.5kg and I consider this quite heavy. I think the only one that I've seen of late that was anything more than that was the radio-controlled ostrich. Maintain a visual line of sight to avoid collisions. It's specifically to avoid collisions. If you're going to fly into something or somebody's going to fly into you, you need to know this so that you can take evasive action. That applies even if you're flying with goggles, you need to have somebody else's competence standing somewhere where they can see that you're about to hit something and tell you so that you can take evasive action. It's above ground level. Now, 400ft is quite a scary way away if you're flying line of sight. But if you've got GPS and stuff, it's worth being aware of that as a limit. And not for aerial work. If you're making money out of it, then different regulations apply and you need to show your competence and give the CAA some money for a permit, probably. So this is the easy way if you just want to fly around with fun. As soon as you put a camera on it, it becomes a surveillance craft. If that camera is used for anything other than just flying it. So if you've got a camera with goggles, that might not be a surveillance craft, but if you record it, it is. Then you have extra restrictions, which are kind of relevant here, because you have to stay 150m away from congested areas if you cannot fly over them. And that also includes any open air assemblies of over 1,000 people, which EMF is. And you have to stay 50m away from vessels, vehicles and structures not under your control. So you can fly around your own house as long as you live in the middle of nowhere. But if it's somebody else's house, you have to have effectively their permission. And 50m away from persons not under your control. If you can shout duck and they do, does that count as under your control? I'm not sure. Right, so now you've got to the point where you can start actually flying. How do you do it? This is quite a common transmitter. It has two sticks on it. Lots of other buttons and knobs, but most of them are irrelevant. These two sticks, this is the most common configuration, but there are others. So check your manual before you fly. So the left stick controls the throttle. You stick it up full power. You really want to hold that down until you are not standing over it. And if you move the left stick sideways, that controls your your axis. So as a reminder, this is your. Your pitch is controlled by moving the right stick forwards and backwards, and your roll is controlled by moving the right stick left and right. This will be quite important when you actually take off. So that's your basic control sorted. Now you get to learn to fly it. The best advice I can give you is to keep the front of it pointing away from you. That way, when you move the stick left, it goes left. When you move the stick right, it goes right. If you do it the other way round, you will get confused. And your very first step is going to be taking off. That's quite easy. You bring the throttle up to where it's about to take off. You give it a bit extra, it lifts itself off. Then you have about three seconds to figure out how to fly it before you crash. So the majority of that is going to be learning how to hover, and landing is just a case of hovering just above the ground, trying not to move and going down a little bit. And here is a video of my first quadrata at EMF 2012, about two days after it's made and flight, while it was still learning how to fly this. So you can't see that very well. Who wants to see this flying on stage? Right, okay. You'll notice I'm leaning over this. I trust my pre-flight checks on this. The stage is going to be nice and clean after this. I should have practised this before the talk. I'll disarm it in a minute. So I just want to finish off with things you could do after that, because this was still flying in stabilised mode. So the next steps are fairly obvious. You add more sensors, you can do more things. You can add a magnetometer to give you an absolute reference for your... so that if you want to fly north, you can fly north, as opposed to just some direction. You can add an altimeter so that you can control how high you are. By you, at this point, I mean the computer, right? And similarly, a rangefinder pointing at the ground gives you more accurate control when you're quite low. And the other one that's very attractive is GPS so that you can tell it, go and fly to those coordinates over there and hope that I got them in the right country. So I'm just going to play you out with... the other thing that you can do to it is to attach a camera. Thank you very much.