Developing a GPS Guided Recovery System

Discussion in 'Rocketry Electronics and Software' started by Winston, Mar 6, 2019.

Help Support The Rocketry Forum by donating:

  1. Mar 6, 2019 #1

    Winston

    Winston

    Winston

    Lorenzo von Matterhorn

    Joined:
    Jan 31, 2009
    Messages:
    6,208
    Likes Received:
    217
    Gender:
    Male
     
    gfunk likes this.
  2. Mar 7, 2019 #2

    rickman2k

    rickman2k

    rickman2k

    Active Member

    Joined:
    Jan 3, 2015
    Messages:
    25
    Likes Received:
    4
    I've thought about this b4, was thinking how to get a GI Joe to pull the chords. I like your servo pull down for a left or right turn but have you ever thought about pulling down on both sides at the same time to feather the chute just feet above the ground for a feather soft landing ?

    the second one shows it real good:
     
  3. Mar 8, 2019 #3

    Chad

    Chad

    Chad

    Well-Known Member

    Joined:
    Jul 23, 2018
    Messages:
    113
    Likes Received:
    17
    Gender:
    Male
    Location:
    Dallas
    that's freaking awesome

    I would have thought an IMU hanging between the chute and rocket would be so noisy it would be unuseable. The accelerometers have to be going crazy the whole way down (same for the compass).
     
  4. Mar 10, 2019 #4

    Handeman

    Handeman

    Handeman

    Forum Supporter TRF Supporter

    Joined:
    Jan 18, 2009
    Messages:
    7,414
    Likes Received:
    75
    Location:
    Stafford, VA
    Check out RC parachuting. The RC guys have been using ram air chutes and bodies with articulated/servo controlled arms to do parachute contests for years. Usually they fly them up with a RC plane and drop the skydiver which is controlled with a different radio. It shouldn't be hard to put a GPS unit in place of the RC radio. If you have a GPS lock when the rocket deploys the unit.

    I've built a unit that is RC controlled to fly a rocket back to the pad. The issue I've had is deployment of the chute. With skydivers, RC and real, they have a known position they are in when the chute is deployed and how it deploys. With a rocket, not so much and the chute is packed in a tube formation instead of a flat back pack like a skydiver. The RC control was easy, getting it to deploy correctly in all the various ways a rocket deploys a chute was a whole different problem. Add the complexity of having a GPS lock on enough satellites to know where it's at, good luck...
     
    Chad likes this.
  5. Mar 10, 2019 #5

    cerving

    cerving

    cerving

    Owner, Eggtimer Rocketry TRF Supporter TRF Sponsor

    Joined:
    Feb 3, 2012
    Messages:
    3,003
    Likes Received:
    192
    Really all you need is a GPS, no compass or IMU... you're not going to be moving so fast that you're not going to keep a fix. The real challenge is the steering... that looks pretty dang difficult to me.
     
  6. Mar 10, 2019 #6

    Chad

    Chad

    Chad

    Well-Known Member

    Joined:
    Jul 23, 2018
    Messages:
    113
    Likes Received:
    17
    Gender:
    Male
    Location:
    Dallas
    What about duel deployment? Would it be easier to deploy a flyable chute correctly after a drogue has been out for a bit and the rocket is in a more predictable state?
     
  7. Mar 10, 2019 #7

    Handeman

    Handeman

    Handeman

    Forum Supporter TRF Supporter

    Joined:
    Jan 18, 2009
    Messages:
    7,414
    Likes Received:
    75
    Location:
    Stafford, VA
    That is exactly what I was attempting, but it would probably be easier at apogee. The orientation is much more random while falling under drogue. The payload points at the ground and up almost to the horizon as things fall. Either way, the biggest problem was the shock cord and nose cone. The ram air chute almost always tangle on the shock cord.

    The easiest solution would be to put the chute in a d-bag and a pilot chute on the d-bag and nose cone, but I didn't like the idea of loosing a nose cone every flight since it would fall at a completely different rate then the rocket and it would go with the wind. If you have to walk miles to get the nose cone, what's the point of having a guided main?
     
  8. Mar 10, 2019 #8

    Handeman

    Handeman

    Handeman

    Forum Supporter TRF Supporter

    Joined:
    Jan 18, 2009
    Messages:
    7,414
    Likes Received:
    75
    Location:
    Stafford, VA
    The system I built was a 7-cell ram air chute. I basically downsized a man type chute or up sized a RC parachute, whichever way you want to look at it. It's 6.5 ft. wide and about 36" deep with 5" tall cells. I used an expert system e.g RC controls.
    I also used two servo so I could steer with one or the other, or slow forward speed by pulling both control lines at the same time like skydivers do to get the soft "step down" type landings. One of the problems was getting enough movement on the control lines to have effective control.
    I also had a deployment charge controlled by the RC so I could use the RC to deploy the chute and didn't have to rely on the apogee/main charges of the altimeter.
    As Tim said at 48:30 into the video, it is a parachute and bad things can happen. It's easy to tangle the chute lines with the shock cords. That was the biggest problem I've had. I just couldn't get a consistent deployment.

    I haven't worked on it in quite a while. There were some nice insights I got from the video, so maybe I'll start working it again. Maybe simplify the whole system and use apogee deployment.
     
  9. Mar 10, 2019 #9

    Winston

    Winston

    Winston

    Lorenzo von Matterhorn

    Joined:
    Jan 31, 2009
    Messages:
    6,208
    Likes Received:
    217
    Gender:
    Male
    Yes, the addition of GPS allows hands-off return to launch site capability even for beyond visual range flights, but for anything within visual range, a very cheap manually controlled system using an RC receiver and servo could be rigged up. I think I've read about examples of that posted here before.

    EDIT: It just occurred to me that, since the definition of what defines an Unmanned Aircraft System which requires owner/pilot registration with the fedgov is anything over 250g flight weight that meets this all encompassing definition - "An "unmanned aircraft system" includes the communication links and components that control the small unmanned aircraft along with all of the other elements needed to safely operate the (aircraft)" - the use of a GPS controlled or manually RC controlled return to launch site system may get the rocketry hobby into the UAS registration requirements unless specifically exempted. NOTE also that even hobbyist RC aircraft with on-board flight controllers and GPS which are capable of and designed for autonomous flight are prohibited from flying beyond visual range.
     
    Last edited: Mar 10, 2019
  10. Mar 11, 2019 #10

    Handeman

    Handeman

    Handeman

    Forum Supporter TRF Supporter

    Joined:
    Jan 18, 2009
    Messages:
    7,414
    Likes Received:
    75
    Location:
    Stafford, VA
    I haven't looked into the requirements, but does it just define it as anything in the air, or does it have to have propulsion? If it does, that might exempt rocket recovery.
     
  11. Mar 16, 2019 #11

    Winston

    Winston

    Winston

    Lorenzo von Matterhorn

    Joined:
    Jan 31, 2009
    Messages:
    6,208
    Likes Received:
    217
    Gender:
    Male
    Anything that has a remote control link or even an autonomous onboard control system, powered or unpowered, falls under the pilot registration requirement for anything over 250g . For instance, RC sailplanes over 250g require pilot registration.

    BTW, assuming that accidental collisions with stuff is the real reason for this RC pilot registration thing, isn't it really strange that anything powered or unpowered, but which is designed and used for free flight (i.e., totally uncontrolled) doesn't require pilot registration? Don't ya' think that something totally uncontrolled would be more likely to accidentally hit something? Not that anyone would build a free flight aircraft this big, but up to 55 lbs would be the totally unregulated weight limit for a free flight hobby aircraft just as it used to be for all hobby aircraft. Kinda' gives away the real reason for RC pilot registration, doesn't it? Nudge, nudge...
     
    Steve Shannon likes this.
  12. Mar 16, 2019 #12

    Winston

    Winston

    Winston

    Lorenzo von Matterhorn

    Joined:
    Jan 31, 2009
    Messages:
    6,208
    Likes Received:
    217
    Gender:
    Male
    Here's the latest. A lawsuit killed the original registration thing, but of course they found a way around that:

    https://amablog.modelaircraft.org/a...-final-rule-for-external-marking-requirement/

    Q: Do only drones and multirotor operators need to register?

    A: Anyone who flies a model that can freely navigate in the air and uses a remote control device (e.g. RC transmitter) is required to register. This includes drones, traditional fixed wing model aircraft, model helicopters, and other remote controlled model aircraft. If you exclusively fly models under a half pound (250 grams or .55 pounds), indoors, control line, or free flight models – you do not need to register.

    Note that fully autonomous aircraft are not mentioned there, but I could swear I read something about it somewhere because that topic interests me far more than human-controlled RC flight. NOTE that such autonomous hobby aircraft also have a remote control system in every case I've read about, so perhaps it is only THAT which triggers the registration requirement by a strict interpretation of the FAA's rules in which case a fully autonomous GPS guided rocket recovery system would not violate the rules as summarized above.

    However, considering what I believe is the real intent of pilot registration as further revealed by the new rule linked to below, I suspect there must be something elsewhere that is not stated in that short Q&A summary above that would prohibit fully autonomous flight. After all, autonomous drones were used to attack Russian forces in Syria:

    Russia Offers New Details About Syrian Mass Drone Attack, Now Implies Ukrainian Connection
    Though no group has taken responsibility, Russian authorities continue to insist the attackers needed outside support.

    http://www.thedrive.com/the-war-zon...drone-attack-now-implies-ukrainian-connection

    ...Novikov focused heavily on the drones’ on-board GPS, which has become an important component of Russia’s allegation that outside actors, including one or more nation states, would have had to have been involved in the attacks at some level. The Major-General said the systems gave them military-grade precision when navigating to Khmeimim and releasing their bombs over the target area.

    FAA Registration, Renewal, and Marking Your Aircraft

    https://amablog.modelaircraft.org/a...gistration-renewal-and-marking-your-aircraft/

    As of Monday February 25, 2019, the FAA’s Interim Final Rule requires drone pilots and model aircraft pilots to display their FAA-issued registration number on the outside surface of their aircraft.

    Note that originally the pilot registration marking could be placed in the battery bay. However, as stated in regulation discussions found in knowledgeable forums, the reason for the external marking only requirement is that first responders don't want to have to handle the "drone" and open it up. What's that tell you about what they're actually worried about? Also mentioned in those forums is that there were over a hundred incidents in 2018 where first responders were called due to a "drone" found on the ground or where there was an intrusion into a prohibited airspace like over a military base.

    Note also that the reason markings were originally allowed inside the "drone" was due to the RC community's fear of someone else who REALLY doesn't like you seeing your pilot registration number, putting it on a "drone," doing something very illegal with it, and then having your door knocked down by LE.
     
  13. Mar 17, 2019 #13

    Winston

    Winston

    Winston

    Lorenzo von Matterhorn

    Joined:
    Jan 31, 2009
    Messages:
    6,208
    Likes Received:
    217
    Gender:
    Male
    I note that unlike in the past, there is no explanation anywhere I can find at the "drone" registration links of what actually REQUIRES RC pilot registration other than any aircraft that weighs over 0.55 lbs.

    Step 1: Register Your Drone

    Even if you're only flying in your backyard, drones that weigh more than 0.55 pounds must be registered.


    https://faadronezone.faa.gov/#/

    https://www.faa.gov/uas/getting_started/

    https://www.faa.gov/uas/recreational_fliers/

    Here's why. Apparently ANY aircraft that weighs over 250 grams (0.55 lbs) now requires pilot registration. They apparently don't even exempt control line (tethered) aircraft any more!:

    U.S. CODE OF FEDERAL REGULATIONS
    Regulations most recently checked for updates: Mar 15, 2019

    § 48.15 - Requirement to register.

    No person may operate a small unmanned aircraft that is eligible for registration under 49 U.S.C. 44101-44103 unless one of the following criteria has been satisfied:

    (a) The owner has registered and marked the aircraft in accordance with this part;
    (b) The aircraft weighs 0.55 pounds or less on takeoff, including everything that is on board or otherwise attached to the aircraft; or
    (c) The aircraft is an aircraft of the Armed Forces of the United States.


    Since model rocketry has its own rules and are never specifically described as "aircraft" that I can find, guided parachute recovery systems most likely would not require registration. But I'd ask...

    -----------

    BTW, here's official proof of the reason for that change to the sUAS pilot registration number marking requirement:

    External Marking Requirement for Small Unmanned Aircraft
    A Rule by the Federal Aviation Administration on 02/13/2019

    https://www.federalregister.gov/doc...rking-requirement-for-small-unmanned-aircraft

    "Members of the law enforcement and security communities have expressed concern that the current rule, which allows registration numbers to be marked in an enclosed compartment, presents an imminent risk of harm to first responders. When responding to a security incident involving an unmanned aircraft, first responders seek to identify the owner or operator. One way to do that is to obtain the registration number of the unmanned aircraft. Requiring first responders to physically handle a small unmanned aircraft to obtain the registration number poses an unnecessary safety and security risk to those individuals, as well as to others in the immediate proximity to the aircraft, because of the potential for the unmanned aircraft to conceal an explosive device in an enclosed compartment (such as the battery compartment), designed to detonate upon opening. Requiring small unmanned aircraft owners to place the registration number on an external surface of the aircraft helps to mitigate this risk because a first responder can view the number without handling the aircraft, or by using other technologies that allow for remote viewing of the aircraft's external surface."
     
  14. Mar 17, 2019 #14

    BLH

    BLH

    BLH

    Well-Known Member

    Joined:
    Dec 6, 2018
    Messages:
    108
    Likes Received:
    26
    I saw this presentation at NARCON. He explains why GPS only doesn't work, why ram air chutes are not optimal and some other interesting points.

    In talking with them, I guess deployment is not that hard either and I think they said they are getting above 90% success rates. That is better than chute releases that seem to be somewhere in the 70% range. They said the key to making any higher performance chute deploy "clean", is pull the nose out first.

    Who knows, still really cool.
     
  15. Mar 17, 2019 #15

    BLH

    BLH

    BLH

    Well-Known Member

    Joined:
    Dec 6, 2018
    Messages:
    108
    Likes Received:
    26
    So does this mean that anyone who fly's Joe Banard's new vectored thrust control from BPS space or using canard fin's like Jim Jarvis is trying has to get this special license?
     
  16. Mar 17, 2019 #16

    Nytrunner

    Nytrunner

    Nytrunner

    Master of Rivets

    Joined:
    Oct 15, 2016
    Messages:
    4,018
    Likes Received:
    614
    Gender:
    Male
    Location:
    Huntsville AL
    An issue someone mentioned at NARCON: programmable landing aimpoints. Default is to return to launch point, but if the option exists to input an Other location, how is that not Targetting?
     
  17. Mar 17, 2019 #17

    BLH

    BLH

    BLH

    Well-Known Member

    Joined:
    Dec 6, 2018
    Messages:
    108
    Likes Received:
    26
    Didn't sound like that was an option. Exactly so it couldn't be used for targeting...
     
  18. Mar 21, 2019 #18

    Nytrunner

    Nytrunner

    Nytrunner

    Master of Rivets

    Joined:
    Oct 15, 2016
    Messages:
    4,018
    Likes Received:
    614
    Gender:
    Male
    Location:
    Huntsville AL
    I must have interpreted this line incorrectly. I guess he's referring to where you launch being the place you "pick" to return to

    upload_2019-3-21_10-28-20.png
     
  19. Mar 21, 2019 #19

    BLH

    BLH

    BLH

    Well-Known Member

    Joined:
    Dec 6, 2018
    Messages:
    108
    Likes Received:
    26
    Yay, that does read odd... Not sure about all that, but I know the guy said that targeting would never be an option. They were even thinking of adding a launch detect scheme that would be required to enable the system and check the landing spot is "close" to prevent someone from turning it on at one location and then, without turning it off, moving to a different location....

    PS: what do the little green triangles mean in the upper left corners by the user logo? It seems your and my posts are only being seen by us? Maybe because we have the "T" word...
     
  20. Mar 21, 2019 #20

    Nytrunner

    Nytrunner

    Nytrunner

    Master of Rivets

    Joined:
    Oct 15, 2016
    Messages:
    4,018
    Likes Received:
    614
    Gender:
    Male
    Location:
    Huntsville AL
    Online now

    I usually have a forum tab up in the background to read at lunch
     
  21. Mar 21, 2019 #21

    BLH

    BLH

    BLH

    Well-Known Member

    Joined:
    Dec 6, 2018
    Messages:
    108
    Likes Received:
    26
    Thanks Rodger. Nonetheless, I hope they get this thing on the market soon and it works reasonable. I want to target my rockets to where the TREE's aren't!!!
     
  22. Mar 23, 2019 #22

    afadeev

    afadeev

    afadeev

    Well-Known Member

    Joined:
    Sep 21, 2017
    Messages:
    692
    Likes Received:
    120
    Agreed, thanks "Wiston" for posting this.
    I finally had time to catch up and listen to the long and informative video, and learned a ton about chutes.


    In my humble experience, chute release has about 99% release success rate.
    Then, 1 out of ~20 times subsequent to JL CR releasing, the chute itself may fail to fully inflate. That could be either due to packing error, or some other weird condition (shroud lines tangled up beyond recognition during ejection, etc).

    If/when this GPS glider-chute product goes to market, it might achieve better results than JL CR, since the glider should be actively flying towards you.
    However, there is one important caveat.
    They seam to have settled on a chute design that can generate 5-7 mph of gliding speed. Most of the times I fly, the upper alleviation winds are gusting significantly above 10mph, usually 15-20+ mph. The winds will simply over-power the glider. As was illustrated by the last flight video and graphs.

    I suppose one could dual-deploy to avoid upper altitude winds, but then you are at the mercy of rocket drifting under drogue. And then the marginal value from GPS glider-chute is precision flying back to the launch pad for from the point of main deployment at 300-800 feet. But by that point, the rocket could already have drifted too far away.

    Either way, the tech is cool.

    a
     
  23. Mar 25, 2019 #23

    Winston

    Winston

    Winston

    Lorenzo von Matterhorn

    Joined:
    Jan 31, 2009
    Messages:
    6,208
    Likes Received:
    217
    Gender:
    Male
    I doubt it for the reasons I pointed out in my post #13 above plus a serious lack of range of that system. It's stabilization system is a novelty, interesting for technical reasons, but very inefficient because of the control system and the mass of its components. Also, with something like the GPS guided return to launch site parachute system covered in the video which started this thread, that's not a threat either unless the bad guys are wanting to blow THEMSELVES up!

    On further consideration I think the measure of anything that might be wise to check with the FAA about would be something that could be used as a remotely controlled or a programmable GPS guided airborne vehicle which could be used as a weapon since, if you spend any time at all looking at the numerous on-line articles of such potential and actual use concurrent with the sudden desire to require the registration of RC pilots in a hobby with an incredibly safe record throughout its history, you will realize that is what pilot registration is all about. Further circumstantial evidence of that is that all RC flying within 30 miles of DC was totally prohibited immediately after the registration requirement became active. That was later lifted.

    So, remotely controlled FPV or GPS guided rockets would be a big no-no if we want to keep such regulation away from our hobby.
     

Share This Page