GPS steer-able parachute

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Haven't checked out the link, but I wonder... To get back to the launch area is kinda like sailing upwind, isn't it? I never did figure out how tacking worked; but long ages ago I was taught that whatever heading you set in an aircraft, you must take the wind speed and direction and add them vectorially to get your actual ground heading. Sure would be nice, though!

Back to work...
 
Check out Hackaday. The author of this project is using the homemade device with an eye to high-altitude balloon drops. Ought not to take some bright young rocketeer much time to figure a setup that works for rockets.

https://hackaday.com/2021/01/07/gps-guided-recovery-parachutes-for-high-altitude-balloons/

Thank you so much for sharing!

The good news is that although it's not the main objective at the moment, I have in mind that it can be useful for model rocketry, and so the system is already designed to be possibly integrated as a coupler bay in a model rocket.

I had started a thread on this forum a few weeks ago, I will post an update on it very soon. : https://www.rocketryforum.com/threa...stem-for-rocketry-and-weather-balloon.162520/

>To get back to the launch area is kinda like sailing upwind, isn't it?

It is, and all the ability to go upwind, is based on the flight performance of the ram-air parachute canopy, the first trick is to have the lowest possible drag, the second trick is to vary the wing loading of the wing, to make it fly faster or slower, so, for example, by loading it to fly at 30km/h, if there is 20km/h of wind, you will always be able to fly at 10km/h against the wind.

However, it works for rockets, but with weather balloons, the wind in altitude while the system is still under balloon is really too strong, that's why we wouldn't try to land at the launch location, but more in a safe zone, more or less on the trajectory of the wind.
 
Unfortunately I can think of some nefarious uses for such a device.
Agreed. Though I think that any villain would have to come up with something that would hold a whole LOT more destructive material. Considering the kind of destruction that could be done with a large RC airplane, I don't think balloon-launched or rocket-launched nefarious devices are going to be used any time soon. (I hope!)
 
Really, though, can't any of this stuff be used for nefarious purposes? We do scale models of ballistic missiles, air-to air & air-to-ground missiles, etc. Criminee, if I could launch out on the playa, I'd be tempted to try n explosive warhead just for the fun of it!

<and everyone backs away slowly> ... oh... would you...?
 
Haven't checked out the link, but I wonder... To get back to the launch area is kinda like sailing upwind, isn't it? I never did figure out how tacking worked; but long ages ago I was taught that whatever heading you set in an aircraft, you must take the wind speed and direction and add them vectorially to get your actual ground heading.
Sailboats have to tack, moving diagonally back and forth in the water, because sailboats can not sail straight into the wind.

Once a model is in the air, forget about the ground as far as affecting what is in the air. The wind direction and velocity is what affects it. Once a chute deploys, from the model's perspective, it is in CALM air, that just happens to have ground moving under it at "X" velocity and "Y" direction (X being windspeed, and Y being direction). Note, for simplicity I'm not counting gusting winds, but a steady wind. People flying in a hot air balloon in a 10mph wind feel like they are in calm air, as they watch the Earth move underneath them at 10 mph.

For a plane pilot trying to navigate from point A to point B using a compass and knowing the windspeed and wind velocity, as well as the planes speed, then yes they need to calculate the compass bearing to use to correct for the wind vector. But for a relatively "dumb" GPS system, it is simply going to steer the model to the designated landing spot (which for R/C multicopters and R/C planes with Return To Home (RTH) GPS, is the same spot where the model was when it was turned on and got GPS lock). It knows nothing of windspeed or wind vector. It just steers straight for the GPS spot. Of course the wind will try to mess with that. For multicopters, they can handle it and fly straight anyway. An R/C plane, it gets trickier since an R/C plane can't "crab" into a crosswind very well, at least not one with a simple autopilot GPS system. So it'll get pushed a bit to the side.

Say there is a strong wind blowing to the East, while the model is for whatever reason 1000 feet north of the landing spot and trying to fly straight due south. The wind will try to push the model to the East, bit by bit. And so the GPS will end up needing to steer a bit to the west as it is flying south. So the flht path might not be straight , but sorta curved. Worst-case the model flght path would curve so much east that eventually it is say 500 feet due east of the landing site, and the GPS is then pointing it directly to the west, and the modle slowly flies westward fighting the wind. Now, one could quibble aobut that with an R/C plane, but pretty much that's what a steerable parachute may need to do. Because gliding sterrable parachutes do not have a really fast glide, and if there is a lot of wind then the chute has to do a lot to fight that crosswind. So this is why I expect a gliding parachute might drift in a strong crosswind until it has curved its path of course enough to approach more straight into the wind.

I will say I've flown R/C gliders in a lot of sucky wind. Well, winds that BLOW. Sometimes, in say a 14 mph wind, and the glide speed is 15 mph. Takes a lot of careful piloting to keep the nose pointed directly into the wind, to fly "forward" at 1 mph. If I tried to "tack", letting it get diagonal to the wind, it would quickly get farther and farther downwind. The good thing with an R/C glider that has elevator control, is that it can be made to glide faster by giving down trim to put the nose down a bit, to pick up speed, at the cost of descending quicker. But that can be well worth it to get the glider to come back to you. Bad thing about steerable chutes is they can't do that, no way to increase their forward "glide" speed (that I am aware of anyway). So it can be very important to know what the glide speed is for a steerable chute, and simply not fly when the wind AT ALTITUDE is likely too close to the glide speed of the steerable chute (note that wind a few hundred feet up is often a lot stronger than at ground level). Oh, I've also flown R/C gliders in wind that's faster than the normal glide speed, so the only way to not go downwind was to use enough downtrim to have more glide velocity than the windspeed (those situations were important contests, or team practices. Otherwise I'd never have flown in winds like that. It's weird to land a "glider" vertically, with glidespeed = windspeed).

And another joker in the deck in that regard is how "stable" can a steerable chute system be, given the dynamics of contorl, oscillations, and the rocket swinging underneath. What I mean is, can it fly steady straight intothe wind, or not. It may be able to "average" flying into the wind, but if it is swaying left and right, then the more it sways away form straight into the wind, the less forward progress it makes, and it may end up drifting downwind as it zig-zags back and forth (depends on the windspeed, gliding chute glide speed, and the losses from zig-zagging instead of being steady straight into the wind).

All that said, I really like the idea of it. The videos I've seen of the Apogee system being tested, is impressive, And if Apogee was not already working on it, it might have been a project I'd look into doing.

So I'm not attmepting to be negative about it. Just that I expect there to be some practical limitations when comes to fliers expecting a GPS Chute system to bring their model back REGARDLESS of the wind.
 
Last edited:
Sailboats have to tack, moving diagonally back and forth in the water, because sailboats can not sail straight into the wind.

Once a model is in the air, forget about the ground as far as affecting what is in the air. The wind direction and velocity is what affects it. Once a chute deploys, from the model's perspective, it is in CALM air, that just happens to have ground moving under it at "X" velocity and "Y" direction (X being windspeed, and Y being direction). Note, for simplicity I'm not counting gusting winds, but a steady wind. People flying in a hot air balloon in a 10mph wind feel like they are in calm air, as they watch the Earth move underneath them at 10 mph.

For a plane pilot trying to navigate from point A to point B using a compass and knowing the windspeed and wind velocity, as well as the planes speed, then yes they need to calculate the compass bearing to use to correct for the wind vector. But for a relatively "dumb" GPS system, it is simply going to steer the model to the designated landing spot (which for R/C multicopters and R/C planes with Return To Home (RTH) GPS, is the same spot where the model was when it was turned on and got GPS lock). It knows nothing of windspeed or wind vector. It just steers straight for the GPS spot. Of course the wind will try to mess with that. For multicopters, they can handle it and fly straight anyway. An R/C plane, it gets trickier since an R/C plane can't "crab" into a crosswind very well, at least not one with a simple autopilot GPS system. So it'll get pushed a bit to the side.

Say there is a strong wind blowing to the East, while the model is for whatever reason 1000 feet north of the landing spot and trying to fly straight due south. The wind will try to push the model to the East, bit by bit. And so the GPS will end up needing to steer a bit to the west as it is flying south. So the flht path might not be straight , but sorta curved. Worst-case the model flght path would curve so much east that eventually it is say 500 feet due east of the landing site, and the GPS is then pointing it directly to the west, and the modle slowly flies westward fighting the wind. Now, one could quibble aobut that with an R/C plane, but pretty much that's what a steerable parachute may need to do. Because gliding sterrable parachutes do not have a really fast glide, and if there is a lot of wind then the chute has to do a lot to fight that crosswind. So this is why I expect a gliding parachute might drift in a strong crosswind until it has curved its path of course enough to approach more straight into the wind.

I will say I've flown R/C gliders in a lot of sucky wind. Well, winds that BLOW. Sometimes, in say a 14 mph wind, and the glide speed is 15 mph. Takes a lot of careful piloting to keep the nose pointed directly into the wind, to fly "forward" at 1 mph. If I tried to "tack", letting it get diagonal to the wind, it would quickly get farther and farther downwind. The good thing with an R/C glider that has elevator control, is that it can be made to glide faster by giving down trim to put the nose down a bit, to pick up speed, at the cost of descending quicker. But that can be well worth it to get the glider to come back to you. Bad thing about steerable chutes is they can't do that, no way to increase their forward "glide" speed (that I am aware of anyway). So it can be very important to know what the glide speed is for a steerable chute, and simply not fly when the wind AT ALTITUDE is likely too close to the glide speed of the steerable chute (note that wind a few hundred feet up is often a lot stronger than at ground level). Oh, I've also flown R/C gliders in wind that's faster than the normal glide speed, so the only way to not go downwind was to use enough downtrim to have more glide velocity than the windspeed (those situations were important contests, or team practices. Otherwise I'd never have flown in winds like that. It's weird to land a "glider" vertically, with glidespeed = windspeed).

And another joker in the deck in that regard is how "stable" can a steerable chute system be, given the dynamics of contorl, oscillations, and the rocket swinging underneath. What I mean is, can it fly steady straight intothe wind, or not. It may be able to "average" flying into the wind, but if it is swaying left and right, then the more it sways away form straight into the wind, the less forward progress it makes, and it may end up drifting downwind as it zig-zags back and forth (depends on the windspeed, gliding chute glide speed, and the losses from zig-zagging instead of being steady straight into the wind).

All that said, I really like the idea of it. The videos I've seen of the Apogee system being tested, is impressive, And if Apogee was not already working on it, it might have been a project I'd look into doing.

So I'm not attmepting to be negative about it. Just that I expect there to be some practical limitations when comes to fliers expecting a GPS Chute system to bring their model back REGARDLESS of the wind.
Thank you George.
A very well written description of the dynamics at play here.
NASA played with an inflatable Rogallo chute for Gemini in the 60’s. I will try to dig up some photos of it.
 
Hi George !
Bad thing about steerable chutes is they can't do that, no way to increase their forward "glide" speed (that I am aware of anyway).

All your text, and your thought seems right to me, except this point. On a ram-air steerable parachute, you can control (in flight) flight speed and flight direction. (Pulling the two brake line at once, you make it fly slower, releasing the two brakes line at once, you make it fly faster, pulling one or another brake line, you make it turn)

Once on the ground, by changing the wing loading (by adding or removing weight), you can also change the minimum speed and maximum speed, on a ram-air parachute, the higher wing loading you have, the faster you go.
 
On a ram-air steerable parachute, you can control (in flight) flight speed and flight direction. (Pulling the two brake line at once, you make it fly slower, releasing the two brakes line at once, you make it fly faster, pulling one or another brake line, you make it turn)

Once on the ground, by changing the wing loading (by adding or removing weight), you can also change the minimum speed and maximum speed, on a ram-air parachute, the higher wing loading you have, the faster you go.
I have never heard of the steering lines being able to extend to allow for faster flying. My understanding is that the lines have a fixed (maximum length) position for normal glide straight ahead. The pilot can pull one line down to steer, or both to flare, or to "dump altitude". You are implying that for normal glide the lines are in a middle position, so that the pilot can extend the length of the lines to fly faster. Can you provide some documentation of that? Otherwise the only way I would think of that working, would be for the lines to be extra-long to begin with, to fly extra-fast, and the pilot needing to manually hold both lines down at half-way for most of the flight to fly at normal speed.
 
When I was last in the States I visited the Udvar-Hazy museum at Dulles. Spectacular. The tour guide there had the claim to fame that he was hang-gliding one day, in the early days of the sport, and accidentally crashed into another glider. Both pilots made it down safely. The pilot of the other glider was actually Mr Rogallo 😂. How embarrassment! True story. Even made it to the newspapers.
 
BTW - it occurred to me that it would be "nice" if a GPS steered parachute system was a bit smarter. If the wind is blowing from west to east, and the rocket weathercocks into thr wind, ejecting far upwind, it might land before it gets to the launch site it is steering for. So, it owud have a tial-wind on landing and subject the modle to al ot more damage risk than if it landed facing into the wind. In my theoretical exmaple before, with a model 1000 feet north trying to fly south in a wind blowing form west to east, I theorizt it might end up having a curved path that ended up approaching more from the east, pointed west into the wind. Well, for the sake of a softer landing, it would be great if the chiute tried to land into the wind anyway.

But that would require more programming for the controller. If the flight controller had to recognize "wind drift" on the fly, that's a pretty complex thing to do. Do-able, with enough programming, but quite a challenge for a hobby product for sale. An in-between way would be for a pre-launch step where when the controller is first turned on, and GPS and compass start to memorize where they are, if the flier rotated the controller / compass INTO the wind first before turning it on. The problems with that would be if the flier forgot to rotate it into the wind, or if the wind direction changed significantly between then and liftoff.

With Ardupilot/Arducopter, the flier can use a laptop to program a pre-set landing pattern, to come in on different approach legs and face in a specific direction to land. But that's not too practical to do on the field. I have done it with some multicopters, and also an electric sailplane, to fly a specific course and land in exact way. But I set that up at home, didn't have a laptop running Windows to do that sort of stuff in the field. For the electric sailplane, where it was landing into an empty plowed field with rough rows, I set up the landing parallel to the rows. I did have a north approach file, and a south approach file, to update the controller depending on which way the wind was blowing that day, though of course rarely would the wind be due north or due south. I never did it much, and the flying was in pretty low wind anyway (I was mostly testing to see if it worked, not using it for regular flying, as part of an idea on doing that "autoland" for R/C rocket boosted gliders that might boost say a mile or more up). Nice thing about that electric sailplane is I could fly it way out to a dot in the sky, flip a switch, and it would fly itself back to the take-off area (using low throttle to maintain a pre-set minimum altitude) and then circle around in a 300 foot circle until I flipped back to full manual control. Or flipped another switch to go from return to home mode (which that was), to landing mode where it then carried out the programmed flight path sequence to land on the imaginary "runway" in the plowed field.
 
Last edited:
BTW - it occurred to me that it would be "nice" if a GPS steered parachute system was a bit smarter. If the wind is blowing from west to east, and the rocket weathercocks into thr wind, ejecting far upwind, it might land before it gets to the launch site it is steering for. So, it owud have a tial-wind on landing and subject the modle to al ot more damage risk than if it landed facing into the wind. In my theoretical exmaple before, with a model 1000 feet north trying to fly south in a wind blowing form west to east, I theorizt it might end up having a curved path that ended up approaching more from the east, pointed west into the wind. Well, for the sake of a softer landing, it would be great if the chiute tried to land into the wind anyway.

But that would require more programming for the controller. If the controlelr had to recognize "wind drift" on the fly, that's a pretty compelx thing to do. Do-able, with enough programming, but quite a challenge for a hobby product for sale. An in-between way would be for pre-launch step where when the controller is first turned on, and GPS and compass start to memorize where they are, if the flier rotated the controller / compass INTO the wind. The problems with that would be if the flier forgot to rotate it into the wind, or if the wind direction changed significantly between then and liftoff.

With Ardupilot/Arducopter, the flier can use a laptop to program a landing pattern, to come in on different approach legs and face in a specific direction. But that's not too practical to do on the field. I have done it with some multicopters, and also an electric sailplane, to fly a specific course and land in exact way. But I set that up at home, didn't have a laptop running Windows to do that sort of stuff in the field.

I think it's pretty unlikely that a steerable chute would land with a tailwind. Even if it weathercocked and went far upwind, it would be flying downwind toward the LZ at a noticeable glide slope. With "normal" weathercocking, it's pretty unusual for a rocket to land far upwind of the launch pad on a simple descent under chute. When you add a downwind velocity vector, you'd expect that it would fly past the LZ and then approach from downwind.
 
I think it's pretty unlikely that a steerable chute would land with a tailwind. Even if it weathercocked and went far upwind, it would be flying downwind toward the LZ at a noticeable glide slope. With "normal" weathercocking, it's pretty unusual for a rocket to land far upwind of the launch pad on a simple descent under chute. When you add a downwind velocity vector, you'd expect that it would fly past the LZ and then approach from downwind.
Yep, that should be the case most of the time. I was using an extreme example to clarify what I was trying to describe regarding crosswind and flying back to the site.

Then there is the flip side of that. The model ejects downwind, the GPS steers it back, and there it is, right over the LZ......500 feet up. So it needs to let itself get downwind to some distance so it can then turn back into the wind and steer to the LZ at an altitude and rate of descent that won't overshoot yet again, but try not to land way short. If it was dead calm, a "dumb" GPS would just spiral and spiral once it got to the LZ. Could be programmed to fly a wide enough circle to minimize spiraling, like say a 300 foot circle (see my reference above to an electric sailplane doing that when it returned, except it maintained altitude), but that means in calm it could land anywhere in that 150 foot radius, where there could be tents or cars unless it's launched from a far-away pad.
 
Yep, that should be the case most of the time. I was using an extreme example to clarify what I was trying to describe regarding crosswind and flying back to the site.

Then there is the flip side of that. The model ejects downwind, the GPS steers it back, and there it is, right over the LZ......500 feet up. So it needs to let itself get downwind to some distance so it can then turn back into the wind and steer to the LZ at an altitude and rate of descent that won't overshoot yet again, but try not to land way short. If it was dead calm, a "dumb" GPS would just spiral and spiral once it got to the LZ. Could be programmed to fly a wide enough circle to minimize spiraling, like say a 300 foot circle (see my reference above to an electric sailplane doing that when it returned, except it maintained altitude), but that means in calm it could land anywhere in that 150 foot radius, where there could be tents or cars unless it's launched from a far-away pad.

When I was noodling this in my mind, I assumed that you'd set a pre-programmed landing coordinate somewhere out in the field so you wouldn't risk landing on the pad. I also had a vague notion that once the chute got to within X' (25?) of the landing coordinate, it would pull one side to descend in a spiral. Once it was outside a somewhat larger radius (100'?), it would turn around and head back toward the landing coordinate. That way it would keep circling back, but slow down over the spot. Alternatively, you could have it pull both brakes to just slow forward progress without turning.
 
When I was noodling this in my mind, I assumed that you'd set a pre-programmed landing coordinate somewhere out in the field so you wouldn't risk landing on the pad.
Well, nearly all multicopter type controllers (and airplane verisons) using GPS to "Return To Home (RTH), set the GPS location of where the model is when it is turned on and the GPS receiver starts getting locks on GPS satellites. When I fly a multicopter that has GPS and I make it RTH, it often lands 3-5 feet from where it was when I turned it on.

The easy way to "trick it" to land someplace else, is simply turn it on where you want it to land, let it get GPS lock, then take it to the pad. But of course you have to keep it on, so need to have plenty of battery life and/or be at a launch where you do not have to wait and wait.

As I referred to earlier, there are more advanced controllers like Arducopter & Ardupilot (for planes) where you can indeed set exactly where you want it to land, but you have to do that using a cable hooked up to computer software to make those settings. Of course for a "home field", where you know exactly where you want it to land time after time, you can use that software I refer to, which uses a Google Earth type of thing, to choose the spot where you want it to land, and set that at home. Or at a new-to-you field, could be done at the field using a laptop.
 
Last edited:
You are implying that for normal glide the lines are in a middle position, so that the pilot can extend the length of the lines to fly faster

That's exactly my understanding, you could load the wing so that it is flying at the speed you want with enough of brake lines, and then be able to fly faster, by releasing the two brake lines at once, and slower by pulling the brake lines a bit more. This will in my understanding, change the centrer of drag and center of lift, and so change the angle of attack of the wing, higher angle of attack, lower speed, lower angle of attack, higher speed.

I think it's called "hand's up" and "hand's down" in skydiving and paragliding.

There are also other ways to change speed in flight, take a look at speed trim and speed bar, they directly pull or release the line of some part of the whole canopy to change I think the angle of attack too.

This pdf already provide some good informations about the subject (I haven't finished reading and understanding it yet) :
https://silo.tips/download/precision-aerial-delivery-seminar-ram-air-parachute-design
 
I think our Mr. Gassaway here must walk into stationary objects a lot... between rockets, quadcopters, gliders, and RC aircraft/rocket gliders/quadcopter recovery he's always looking at the sky! ;)

Thanks, Gerorge, and the rest of you for a nice, detailed discussion. 'Way back in high school I had taken and passed my private pilot ground school through AFROTC, but it never occurred to me that the forward glide speed of the 'chute or parawing could exceed the wind speed and thus it would be possible for a downwind rocket to get back home.
 
I know this is digging up an old thread, but I just drove home to Denver from Salt Lake City (a 7.5 hr ride, give or take). I did the driving, so I was mostly thinking about rockets. I went through half a dozen iterations of wing design that could be folded up or something inside a rocket like a scissor wing glider or similar.

Then I remembered steerable parachutes! I've got some, what _I_ think are fun ideas & things to try, but I've got a lot of other things I need to figure out first, so we'll see!
 
Then I remembered steerable parachutes! I've got some, what _I_ think are fun ideas & things to try, but I've got a lot of other things I need to figure out first, so we'll see!
Just make sure you comply with the FAA maximum 400 ft. altitude for UAV flights. That kind of killed most talk of Return To Home systems for rockets which is why the thread is so old.
 
So a parachute is now a UAV? not sure about that , https://www.apogeerockets.com/Gliding-Parachutes
I guess that is why Apogee doesn't include any RC controls with any of those chutes. It's on the user to build the control system and to comply with federal regulations.

From what I understand, if it's self controlled unmanned flight or remote controlled, it falls under the 400 ft. max altitude limit the FAA has, whether you call it a UAV or something else.

Of course, anyone outside the US is going to be dealing with completely different laws and that may be the market Apogee is going for.
 
Back
Top