e this rocket that I built has an active control system it has a custom flight computer in its nose and uses these four aft fins to steer itself through the air this rocket has flown five times now but only two of those flights are up on the channel today we're going to talk about the last three launches flight tests 3 four and 5 all of these launchers were testing Waypoint guidance we're going to give the rocket a in the air and it's going to try to fly to it but before we talk about any of
that we need a little bit of background what am I doing why am I doing it and what happened on those first two test flights I had two goals for the diamond X Program first was to show full aerodynamic control of a model rocket and to improve over some of the previous rockets that i' built both Diamond 1 and Diamond 2 used canards to try and steer and this layout has a number of problems just like the a9x diamond X changed to a tail control design to solve some of these controlability issues my second goal
was to demonstrate some unique types of guidance steering the rocket to way points or a paths in the sky doing these modes would be a huge technical challenge but they would also let me put some of that academic control theory into practice in real life the end goal for both the Waypoint and path following guidance modes is to try to fly the rocket both up wind and down range to help the rocket recover to a safe landing site under its parachute so what has this rocket done so far flight one tested the Rocket's Oiler angle
control system I commanded the rocket to fly straight up and included a 45° roll program this worked almost flawlessly and a video for this flight is out on the channel Flight 2 was all about testing Diamond X from the atlas hot launch Silo the rocket was loaded and held by the metal rails in The Silo and its rocket motor was ignited this is some of the coolest launch footage I have ever taken in the flight the rocket executed a pitch program pitching down to 20° below vertical and back upwards this worked pretty good but it
could have been more maneuverable especially at high speed this launch also damaged The Silo so that entire thing is going to have to get rebuilt and that leads us to today in flights 3 4 and five okay so before we talk about these last three flights we need to talk about some background on what these Rockets do and how they work on board Diamond deck there are three important systems they are the guidance system the navigation system and the control system as I've discussed before the navigation system tells you where you are the guidance system
tells you where you want to go and the control system actually gets you there now I refer to these as guided Rockets where others will describe them as a rocket with an active control system to be very clear these are kind of the same thing if your rocket has an active control system of any type it has to have some guidance system even if it's very simple telling the rocket where to go most amateur rocket guidance systems are trying to control the Rocket's attitude or where it's pointing the most common one of these and probably
the safest is point the rocket straight up this keeps the pointy end up the Flaming ends down and gives you the best chance of safely recovering your rocket there are also some very talented people out there who have landed model rockets propulsively and this is a very very difficult guidance and control problem questions like when do you ignite The Landing motor and how fast should you be going at a given altitude are things that the guidance system has to figure out a lot of the work that I do on this channel is implementing interesting or
novel guidance systems Diamond de flights 3 four and five tested one of these new systems now confusingly Diamond X actually has three different guidance systems that are active at different times during its flight so to help all of the flight footage make sense we're going to go over what those three different systems are how they work and what triggers the guidance mode to change in Flight guidance modes one and three are both variations on Oiler angle guidance this is a type of attitude guidance system that takes pitch yaw and roll set points and tries to
have the rocket achieve them the first mode to talk about is launch Direction hold on launch the rocket isn't going fast enough to do any fancy guidance modes and we also need to calibrate some parts of our control system when we're trying to fly to way points this is the mode that's active as the rocket leaves the rail it tries to point the rocket straight up and get its roll under control to change guidance modes the rocket checks its roll angle and its roll rate if those two values are within certain thresholds it means we've
got the rocket stable and we can transition to Waypoint guidance before launch the rocket is assigned a waypoint these are coordinates relative to the launch site and include a position and an altitude I can choose a waypoint that is both upwind and down range both to shorten our recovery distance and to keep the vehicle away from all the people at the launch site this Waypoint track is assigned remotely via the data link without a track assigned the rocket will fly a pre-programmed pitch yaw and roll profile that's stored on the onboard SD card if I
have assigned a waypoint the rocket will try to fly to that Waypoint instead diamondx tracks its position with its onboard inertial navigation system and it tries to fly as close as possible to this Waypoint the rocket Compares its position and its velocity to the position of the Waypoint and generates lateral acceleration commands several other pieces of guidance and Control software Translate this into fin deflection angles when the fins deflect in pitch or yaw it rotates the rocket to some angle of attack this makes the Rocket's body generate lift and this lift is the force that
causes our lateral acceleration when the rocket detects that it's pass the Waypoint it changes guidance modes for the last time to guidance abort Oiler this is almost identical to the launch Direction hold mode trying to point the rocket straight up and control its roll unlike launch Direction hold though this mode is permanent in the rocket doesn't look to change modes again for the rest of the flight after mode change the rocket will also start reporting its closest approach or its Mis distance over the data link so even if we lose the rocket we can tell
if the guidance and control system worked we point the rocket straight up at the end of the flight to make parachute deployment easier we want to deploy our parachutes while the Rocket's going as slow as possible to minimize the shock loads of the parachutes deploying into the Airstream as we approach our apy our Vertical Velocity goes to zero but our horizontal velocity stays constant if we're going too fast when the parachutes deploy the shot cord or parachute attachment might snap the abort Oiler guidance mode tries to minimize that risk by keeping a rocket traveling straight
up at the end of the flight now there's actually a second reason the rocket might go from Waypoint guidance to abort guidance if there's any mistakes in the Waypoint guidance algorithm the rocket could spiral out of control and come caring back down towards the ground obviously this is very unsafe so Waypoint guidance has a built-in fail safe if the Rock et's pitch or yaw angle ever exceeds 40° the safety system will automatically switch the rocket to abort guidance and will stop trying to fly to our way point now I'm never going to set the Rocket's
way point at such a low angle on the horizon so the only time this system should trip is if there's some fault in the guidance system right this is foreshadowing for later the last thing I'll mention is that you'll see the control system turn on about 20 m in the air this gets the rocket up and off the rail before any of it start to move when we watch the flight footage you'll also be able to see what guidance mode the rocket is in when those modes change you should be able to see the Rockets
Behavior change okay that's enough diagrams let's talk about these last three rocket flights flight three was my first attempt at doing wpoint guidance a waypoint track was assigned up wind and the control system was checked we got ready to launch 5 4 3 2 1 [Music] the rocket lifted off in launch Direct C hold mode and stabilized its role it transitioned to way point guidance successfully and did absolutely nothing it just followed a ballistic path straight up until it was above the wayp point and then it transitioned to aort guidance and recovered successfully so it
turns out I had forgotten a single line of code and this had paralyzed the entire guidance system and left the rocket commanding zero lateral acceleration I had done some last minute tweaking to the guidance algorithm and didn't ground test that so that's on me I guess well okay so the Waypoint guidance didn't work but all the mode transitions worked really well so we did a little bit more ground testing fixed that problem and then it was time to try again flight test 4 was a very similar setup a waypoint 40 m down range and 650
m in the air was assigned via data link given that at this point the whole Waypoint guidance and acceleration control systems were totally untested success here is getting within like 10 met of the Waypoint I would be ecstatic if we got that close 5 4 3 2 1 [Music] [Music] oh [Music] [Music] [Music] there it is out there it reported a 3 m m distance from its Waypoint let's go see if that's real that would be nuts if it was true its actual Mis distance was only about half that far after liftoff we can see
the control system enabling and launch Direction hold getting the vehicle under control at 127 m in the air it transitions to Waypoint guidance and it pitches down range the distance to the Waypoint comes down with acceleration control working really really well 6 seconds into flight at 70 m/ second Diamond X passes the way point 15.6 CM off the Rocket's right side after passing the Waypoint the guidance mode changes to a board Oiler and Diamond X pulls up to vertical at 828 m in altitude 12 and a half seconds into flight the rocket reaches apogee and
deploys its parachutes okay so this flight had a Mis distance of 15.6 CM that is an order of magnitude more precise than I was expecting it to be but does that mean that diamond X has a 15.6 CM Precision how do we even measure the Precision of a vehicle with a guidance and control system so this topic is actually very complicated but the short answer is is no we can't say that diamond X has a 15.6 CM precision and we can't authoritatively say any other number either Precision is a probabilistic process and our understanding of
that process is limited by statistics as a probabilistic process a Rocket's ability to fly along a path or to a certain point is a variable with some distribution the rocket may go here or here but it's most likely to go where the probability is the highest the width of this distribution is related to our Precision if this is a normal distribution we can describe its width with Sigma values or with standard deviation the lower these values the more precise the rocket is but On Any Given flight it may have any Mis distance another way you
may see Precision described is with C or circular error probable in this case this value describes the radius of a circle around our way point if we launch diam index 100 times it would pass the way point inside of the circle 50 of those times and it would pass pass the way point outside of the circle the other 50 again the smaller the circle the more precise the rocket is no matter whether you're measuring with circular error probable or standard deviation or any other probabilistic metric you need to know what the value of that metric
is to find these values we have to collect data we need to launch the rocket a whole bunch of times collect some Mis distances and then do some math on that data set this math will also tell us how confident we are with our estimate of this metric because I have exactly one data point there is no math that I can do that will tell me really how precise this Rockets control system is I'll just have to keep flying it keep tasking it with way points and generating Mis distances and then I'll Circle back to
this whole Precision thing when I have more data so this flight was very precise we'll just keep launching it and see what [Music] happens didn't we have a third rocket launch to talk about [Music] okay so flight test 5 a little bit of a good news bad news situation I did this launch after flight test 4 hence the name uh and I knew that flight test 4 had gone really well so I tked the rocket with a waypoint 80 M up wind 650 m in the air this would give the rocket a little bit more
of a challenge make it maneuver a little bit more and we really test its guidance and control system this is also the second Diamond X airframe this is the one we've been building in that build series and so this rocket has extra Raceway wires to fly with a first stage booster this was just a Shakedown flight testing its hardware and it software on a single stage flight do think is clear Laing in five 4 3 2 one [Music] anybody else noticed how the rest of the rocket is suspiciously far away [Music] [Music] [Music] this flight
suffered two unrelated failures the first was this carbon fiber 3D print this is the wishbone piece that connects our parachute and shot cord to the avionics assembly this part broke when the parachute deployed and so the entire avionics assembly fell from over 2,000 ft this is made out of a very strong carbon fiber polycarbonate uh but unfortunately it wasn't printed at 100% infill like it was supposed to and you can see that it cracked right in the middle of one of the voids in the infill so that was on me I should have printed this
100% infill And I didn't so this part was weaker than it should have been this is also a part that a nylon might be really good at because it's more shock resistant and this is primarily a shock load that it sees when the parachute deploys or when the shock cord snaps toau luckily though it didn't really take much damage the flight computer was actually still on and transmitting Telemetry when I found it and the battery was just kind of dangling out the front of it remember when I said this in the airframe assembly video the
weak links that I designed into these parts are like the arms right here these are very kind of weak so if the rocket hits down sideways the battery will kind of crack off inside of the nose cone and that should take some energy away from the impact well it looks like that's exactly what happened the battery mount broke right on these arms right where we're supposed to the battery cracked off and it was still connected to the fly computer everything else is pretty much damage free so all I have to do to fly this guy
again is reprint the battery Mount reprint the nose and reprint our Wishbone that attaches us to the shock cord and this rocket should be ready to fly again the second failure was its guidance system this wasn't really much of a failure because the rocket did exactly what it was supposed to do but the guidance system ended up ab boarding before the rocket reached the way point okay remember that foreshadowing thing and how I said something would come up later it's now later on this flight the rocket had a really tough time controlling its role flight
Four's Waypoint guidance enabled at 135 m in altitude but flight 5 is enabled at 400 m this meant that the vehicle was already much higher and it needed a much more aggressive maneuver to try to get to its Waypoint as soon as Waypoint guidance enables the rocket pulls almost seven GS before tripping that pitch angle limit that I talked about earlier and reverting to abort angle guidance so I can do three things that will help fix this first I can enable the control system at an even lower altitude this will prevent that big initial Ro
from occurring and we should stabilize our Ro quicker in the flight second I can relax the roll limits before Waypoint guidance enables and this will again help the rocket start guiding earlier lastly I can do a better job of aligning the Raceway and fins before launch and that'll reduce these adverse roll moments that we get so flight four didn't guide successfully but this was a good test of that pitch angle limit and that worked wonderfully so that's a huge safety system to have on board I feel a lot more confident going into the next flights
that I know that that safety system works speaking of safety I wanted to take some time here at the end of the video to talk about how you can do rocketry with or without a control system safely and without getting into any trouble rocketry is actually a pretty safe hobby especially compared to something like motorcycle riding but this doesn't mean that there aren't any risks in fact there are a lot of risks rocketry is safe because we study and we mitigate those risks we have strict safety rules and we follow them a lot of these
rules don't actually come from the government they come from Club organizations like Nar and triple in the United States if you're outside the United States your local rocketry organization certainly has their own safety code as well it's really important for you and everyone else that everyone is following these rules understands the risks and is taking good steps to mitigate them if you're flying a rocket with a guidance and control system you need to take some extra safety precautions here in the United States Nar has some recommendations including increased standoff distances but a lot of the
safety is actually going to come down to you as the engineer you need to design a system that is redundant where it needs to be that fails safely and that's been tested on the ground your guidance or control system failing during a bench test is really frustrating these Systems Failing while your rocket is flying through the air can get people hurt a big piece of reducing risk is modeling and stimulation which I have a whole video dedicated to coming out next lastly I wanted to mention not breaking any laws I have triple and quadruple checked
and I am not aware of any laws in the United States that would stop you from building an active control system for your rocket and flying it that being said I cannot speak for any other countries if you are outside the United States please check your country's laws to make sure you aren't breaking any of them if your Rockets start to steer themselves around even if you are allowed to build Rockets with guidance and control systems obviously you need to be very responsible with them doing something Reckless or unsafe with a rocket intentionally or otherwise
is a pretty good way to get yourself into trouble like prison trouble and if you're ever not sure where the line is that's probably a good time to not get any closer to it there's also this confusing mess of export control restrictions which will vary by country many of these rules are vague intentionally so and and this is one of the reasons why I won't share a code or CAD for these projects online or sell them to people in many respects the line between a rocket and a drone continues to get blurrier so the best
advice I can give you is to learn the rules for wherever you are follow them as closely as you can and always keep safety at the front of your mind okay we can now task Rockets to waypoints what is coming up next we're going to keep flying Diamond X I really want to try some further upwind way points to try to keep those recovery distances short we've also been working through booster integration we should have our first diamond X block 3 flight with a booster in the next few months you guys may have seen me
posting some stuff about Hatchet this is the rocket SL drone thing I'm building with a ducted fan hatchet's being put together right now and I've got its propulsion and control assembly prototype built so if you're interested in that one or any of these future projects make sure you're subscribed and I'll see you guys in the next one [Music]
