I am currently building a rocket with glide return (no on board controller) and am thinking about moving to an autonomous controller for the glide return once I have everything working.

I have no experience with UAVs, but the reading I have done suggests that there could be problems with accelerometers due to the high launch acceleration.

Does anyone here have any experience with Ardupilot in a rocket launch envoronment?



You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –


  • Moderator

    Firstly, thank you Doug for all the information and clear thought pattern you have there. (:

    Secondly, Robert, if you want to talk to someone that had sent something to innerspace etc etc, talk to Monroe.  He'll talk your ear off and offer lots of great advice. 

    However, the rules and regs you need to know about doing this in Australia have been outlined in the group I linked to earlier.  CASA are actually very friendly towards our endeavours and like talking about it.  I have some contacts at CASA you can call, they are great and will offer you any guidance that you need. 

    However in a snapshot -

    • Model Rocket - Means any rocket weighing not more than 1500g which is propelled by one or more rocket motors producing not more than 320Ns total impulse; that contains a device for returning it to the ground in a condition to fly again; whose structural parts are made of paper, wood or breakable plastic and which contains no metal as structural parts.
    • Rockets are excluded from the vast majority of CASA regulations applied to other aircraft (you can exceed the 400AGL limit to other model  aircraft except around aerodromes).
    • I will always use a recovery system (*cough* APM *cough*) in my model rocket that will return it safely to the ground so it may be flown again.
    • And my favourite : Except for insects, my model rocket will never carry live animals or a payload that is intended to be flammable, explosive, or harmful.  Lol.

    Here will help you and check out Appendix D for the guidelines you'll need to follow.


    • Hello

      I am doing High power rocketry (Tripoli L2) and I am currently trying to build a rocket that can glide back as close as possible to the launch using a paraglider. I have successfully built a rover the hard way with a GPS, a compass and an Arduino nano, it more or less goes back to a memorized position and now I would like to do the same for my rocket. Looks like the APM hardware is what I need.... I was wondering if someone had already done some code to modify one of the vehicule to do it? if not which one doing do you think would be the best?

      Note that I am not trying to do any rocket guidance during the ascension of the rocket which is not allowed for obvious reasons



  • After more reflection and thought, Gary has a good point in that none of the major organizations explicitly prohibit a control system. The guidance system referred to in both NRA and TRA codes means the launch rod/rack/tower. The frowning is usually assumed by the folks that think a Model or High Power rocket would could make a WMD. By the limitation on the manufacturers, this is not probable.

    To more specifically focus on Robert's initial question: Call it an R/C Rocket Boost Glider and he should be fine.

    The aerodynamic launch forces are extremely high for an R/C RBG. Control surface separation and shearing is a common event at times. Taking that into consideration AND the fact that the boost phase of Model and High Power rockets is limited, I do not think an APM would be considered a guidance system, but a recovery control system.

    Robert clearly states that his intent is for a glide return. There is no reason he cannot use an APM to enhance the recovery of his RBG.(Not intended to be a legal opinion or counsel)

    Back to his essential question:

    Does anyone here have any experience with Ardupilot in a rocket launch envoronment? (sic)

    An experiment could be done with APM as a payload, oriented with the 'Front' toward the nose of the rocket. Turn APM on and let it record the accelerometer and gyro data -- heck you get barometric too! Recover it, pull the data log, and see what happened.

    APM code could probably be stripped down just for such an application -- call it Ardupilot Payload.

    Possible features:

    • Boost phase profile actions programmable -- air sample collection
    • Add on electronics to sample wind shear at altitude
    • Pictures! Don't forget to take pictures!
    • When it senses no movement, i.e. on the ground, start the 'Here I am' mode to get attention
    • Go look at what other's have done with payloads and adapt APM to that

    Good Luck Robert Hart and make sure to update progress on the build/project.


  • More information as I run this idea to the limit of my curiosity...

    FAA document: (suborbital emphasis -- you go orbital, international laws apply)

    Supplemental Application Guidance for Unguided Suborbital Launch Ve...

    -- note the phrase 'unguided' is prominent.

    A lengthy discussion of Experimental Permits for Reusable Suborbital Rockets from the Federal Register. The TRA is mentioned well into the document.  Tip: Use CTRL-F and search for 'guidance' to see the relevant sections about those systems.

    Another interesting report, slightly dated: Suborbital Reusable Launch Vehicles and Emerging Markets

    The question you pose Robert has been asked, somewhat earlier in THIS post... with similar responses.

    Another discussion on guidance systems HERE. Good points regarding how the project is defined and certified.

    I'm done for now..


  • Rocketry, in the US, is divided into two camps: Model and High Power.. from the NAR website:

    Where Is The Line Between Model and High Power Rocketry?

                A rocket exceeds the definition of a model rocket under NFPA 1122 and becomes a High Power rocket under NFPA 1127 if it:

    • Uses a motor with more than 160 Newton-seconds of total impulse (an "H" motor or larger) or multiple motors that all together exceed 320 Newton-seconds
    • Uses a motor with more than 80 Newtons average thrust (see rocket motor coding);
    • Exceeds 125 grams of propellant
    • Uses a hybrid motor
    • Weighs more than 1,500 grams including motor(s); or
    • Includes any airframe parts of ductile metal.

    NFPA is the National Fire Protection Association -- they handle the high power rocketry rulings.

    APM as a 'payload' may be considered OK and if the payload happens to be on a vehicle that separates from the lift vehicle, this is probably a non-issue with regulating authorities and organizations.

    Again, you have to be specific what your intent and application will be doing. Below I have shown the NAR site and the section that discusses R/C RBG - Radio Control Rocket Boost Glider vehicles. One reason the NAR has definitions of the various types of vehicles is for competition purposes.

    Resources that may, or may not assist you in your research:

    Tripoli Rocketry Association -- High power rocketry certification and education (They sponsor the 'LDRS' meetings)

    Wiki Article TRA and also links to Australian version -- interesting citations of lawsuit with BATF.

    National Association of Rocketry -- the oldest US based organization. Education and Science oriented. Notice who is on their site at the time of this post...


    ...and I also highlighted the portion of the NAR safety code with a link to the RC Rocket Glider codes.

    The NAR has two sections, 'Model Rocket' and 'High Power Rocket' with variations in the safety rules/guidelines.

    Like the AMA, NAR maintains these regulations to make the government agencies happy AND to make insurance companies happy regarding liability during competition and general 'fun flying'.

    Monroe is very interested in suborbital and orbital vehicles. These almost always have some form of payload that is programmable and may have C&C (command and control) capabilities.


  • Moderator

    Hi Robert,

    Look up Monroe Lee King (Team Prometheus) and have a chat to him.  The rules/regs for Australia have also been outlined in the Ardupilot Space Program group so I suggest you look there too.

    CASA will need to know about it, but check out www.casa.gov.au and then post any questions in the Ardupilot Space Program Group. (:

    PM me if you like.


  • Robert,

    Assume Nothing even in AUS , as i suspected you are into the LDRS type of engines . I in your place would contact the AUS version of the FAA and the CSRO(if i remember it right ) about your project and it`s legality in your country . 

     Here LDRS shoots take place in the middle of no where and requires FAA approval and the issuing of a temporary closure of the airspace above and around the launch site for the days of the shoot ,now our middle of nowhere is nothing like the Outback you have to play in :-)

    I like the ideal  by the way it`s just not legal here in the US .

    As for G loading on the electronics you should mount them so the boards do not flex, edge on to axis of G loading  this will stop board failure due to trace and component connections . The accelerometers  i honestly cannot say if they  can take it  but i suspect they can , our military has been developing Excalibur a guided 155 mm artillery round that is GPS guided 

    and has survived a double charge firing (16,000 G`s ) and hit is target within 3 m at a 15 * offset from true aim  at 35.2 km distant target 

    Yes they were mil grade components not consumer grade but gun tube acceleration G loads far exceed rocket launch loading by a factor of i guesstimate 1000-10,000 times .

    The max G loading of the Shuttles (STS ) were less than 10 G`s and everybody who launches satellites has to deal with  it  too  and only a very few satellites make it to orbit Dead on Arrival .

    You are basically wanting to build a Shuttle that never comes close leaving the atmosphere and i believe you could do it with consumer grade components , heck the modern computered car handles high G crashes (much higher than your launch ) and the "Black box " records the high G deceleration with consumer grade components (  accelerometers ).

    Good Luck to you  it would be fun ! 

  • Gary McCray 

    "I suppose that under some interpretation of the Federal Governments "Destructive Device" laws adding a real time controller to a rocket might be seen as potentially dangerous."

    Once you get above D sized engines into LDRS (Large Dangerous Rocket Ships) above off the shelf kit level and engine size  you are in the realm of ballistic missiles ,yea they maybe home made hobby rockets but you have to have a license (Federal full background check ) just to buy ,posses and store the engines .  Put a guidance package in one and you have crossed the line into manufacturing a guided missile that could be used as a weapon of mass destruction and they will treat you accordingly, 3 squares a day ,free room and board for 10 years minimum . They Keep a Very close eye on the LDRS folks.

    They are limited to one way telemetry (Down only) , data recording only on board and timer`s/Altimeters for recovery deployment, no exceptions !!

    NAR does not mention it because they do Not want to put even the ideal out there !!

    Rocket, guidance package, do not pass go ,go straight to jail for unlicensed manufacturer and possession of a DD !! Oh you can get a stamp for  a hand grenade but once you pop it you have to file a disposal form and your stamp is only good for that one only!!

    It`s not an i suppose situation ,They do !!

    As for citing the statute sorry i cannot but i know it to be True !


  • Are you considering this vehicle a rocket power glider or only a rocket that glide returns?

    I ask because of the implication that could be interpreted as a guidance control for a model rocket...something frowned upon.

    If only a glide return recovery scheme I guess the first order would be to have the APM (I am assuming you plan to use APM) 'sleep' during the ascent/boost phase. Once apogee was reached, or just prior to it, you could wake the APM and then assert your control.

    Research could be done on the max impulse the sensor could withstand before failing. If hit hard enough, anything will break.

    In the case of black powder, and probably a composite fuel, there is a slight 'soft start' of the acceleration that could buffer the event. Solid fuels are close to step-wise functions but still the vehicle takes a finite amount of time to reach max velocity.

    An interesting idea/problem case.


This reply was deleted.


DIY Robocars via Twitter
RT @a1k0n: Starting on a rough guess of the track centerline as an SVG. Then I use svgpathtools to generate equally-spaced points w/ direct…
34 minutes ago
DIY Robocars via Twitter
RT @a1k0n: This rules. My existing code isn't going to work *at all* -- the track planner and precomputed map are out. However, it forces m…
34 minutes ago
DIY Robocars via Twitter
RT @chr1sa: Our new @DIYRobocars track at @circuitlaunch is fiendishly hard & crashtastic fun. Combines hairpin curves with an intersection…
2 hours ago
DIY Robocars via Twitter
The May @donkey_car newsletter is out, with news about release 4.2, next events and more https://donkeycar.substack.com/p/may-newsletter
19 hours ago
DIY Robocars via Twitter
DIY Robocars via Twitter
DIY Robocars via Twitter
RT @breadcentric: Bingo card for AWS DeepRacer Finale, starting in 10 minutes on https://www.twitch.tv/aws ! #AWSDeepRacer #DeepRacer #Machin…
DIY Robocars via Twitter
RT @NVIDIAEmbedded: It's #NanoFriday - the RB-0 uses the same suspension concept as #NASA's newer differential-bar rovers. This educational…
DIY Robocars via Twitter
RT @chr1sa: On May 22, we're returning to in-person AI @DIYRobocar racing at @circuitlaunch. The Amazon @awscloud DeepRacer team will be pr…
DIY Robocars via Twitter
RT @breadcentric: On my CV: Hobbies: Training bananas to race on tracks #AWSDeepRacer #DeepRacer https://t.co/MKe14hNyux
DIY Robocars via Twitter
RT @breadcentric: See how the April AWS DeepRacer races have ended and a couple bits of news: https://blog.deepracing.io/2021/05/09/aws-deepracer-league-2021-update-11-end-of-april-special/ #AWSDeepRacer #Machin…
DIY Robocars via Twitter
RT @sunilmallya: Representation Learning +Instance Transfer to learn new reward functions along with advantage based filtering of new exper…
DIY Robocars via Twitter
Apr 27
DIY Robocars via Twitter
Apr 27
DIY Robocars via Twitter
RT @f1tenth: Sliding (autonomously) into the weekend like ... 🤖😎 #f1tenth #robots #AutonomousVehicles @OpenRoboticsOrg @NVIDIAEmbedded @Aut…
Apr 25
DIY Robocars via Twitter
RT @chr1sa: One of the problems with autonomous car racing is that watching software drive is not a very exciting spectator sport. To help…
Apr 25