GPS audible glide angle

[crwper 7]

Hey all!

For the past couple of years, I've been working on a system for providing audible glide angle info to a jumper in freefall. I'd like to put a couple more jumps on it before I put it out here, but will likely post what I've got in a week. In the mean time, I can't help letting the cat out of the bag a bit early.

What you will need: I am using a Garmin GPS 10 and Palm Tungsten T. Any Palm with Bluetooth capabilities will work at the moment, but until I make some changes, the Garmin GPS 10 is required. One of the first things on my list is to make this work with any Bluetooth GPS, so if this interests you please let me know. I am also using a pair of ER-6i earphones, which are easy to hear in freefall. If you have something else you can hear in freefall, that will work, too.

How it works: Wear the GPS as usual. Connect to the GPS using the software before exit. In freefall, it calculates your glide angle using velocity data from the GPS, and beeps like a vario in your ear. Higher pitch/rate means better glide angle. You set the range, so this is useful for trackers as well as wingsuit pilots. At a preset altitude, the thing stops beeping, but logs until you pull. When you land, the software shows you a profile of your jump. You can click on the profile to get specific values.

What I'm working on: First, compatibility with other Bluetooth GPS units. Second, additional displays. Horizontal and vertical speed as well as flight paths are in the works.

What's waaay on the back burner: Sync data to a PC. This is not a high priority for me, since the main objective here is in-flight feedback with some analysis functions after the jump.

Hardware requirements are pretty simple. I've experimented with wired systems, and it's a nightmare, so I'll probably keep this one wireless. The GPS is worn as usual. The Palm can be kept in an inside pocket or belt. The only wire is for the earphone, going from the Palm to your ear. This setup is very simple and really quite easy to use.

The software will be completely free. If you're interested in the source code, that's also free. I've tried to make the software as easy to use as possible. It's not idiot-proof, but I'd definitely say it's up to commercial standards, so it's not like you have to have a degree in computer science to play.

I'd like to hear any suggestions you might have, and I'd be really stoked if anyone out there is interested enough to give it a shot themselves. Let me know!

Michael

[Waldschrat 0]

First: Very nice idea!

This could save a lot of money for beginners if they have the hardware.

Question: Your are not in trouble with the pressurization with this plugs in your ears?

Edit: I hope you add my garmin foretrex in the next time.

[bofh 0]

A guy (Björn Alving) at my dropzone (FKCG, Sweden) has developed his own hardware for this. He has built a tiny box containing a GPS receiver, antenna, LCD display, batteries, some buttons and a serial port for connecting it to a computer.
One can get some basic readings from it in the display and more extensive readings by connecting it to the computer. He has the box connected to the side of his helmet.

His main problem with it is how it should be mounted so more people can use it. It needs to be easy to hold it up next to the window on the plane, for the GPS to get a fix before exit.

[Waldschrat 0]

Klaus Rheinwald has shown me in Hartenholm this tiny GPS-Receiver with Bluetooth and USB.

http://www.wintec.com.tw/en/product_detail.php?pro_id=41

In a Cessna 208 this device receive the GPS satellites without interruption.
My Foretrex 201 badly not.

[Costyn 0]

Very cool man!!
I've been thinking of making something simliar for a long time now, but dont' really have the money for a PocketPC or Palm.

Does the Palm have an USB interface? in that case it might be possible to hook it up to a GPS with serial interface (like ForeTrex models), and output realtime data in NMEA form? I'm just speculating here, I have a ForeTrex 201 and it outputs positional data, should be able to use that somehow.
Anyways, very cool project. And cool that you're making it open source. That kicks ass!

Costyn van Dongen - http://www.flylikebrick.com/ - World Wide Wingsuit News

[Costyn 0]

Quote

In a Cessna 208 this device receive the GPS satellites without interruption.
My Foretrex 201 badly not.

That is pretty strange. The 201 is a fine unit. Were you both on the same plane sitting next to eachother and his worked and yours didn't?

The best place to sit in the 208 (Caravan) is next to the pilot. Alternatively, sitting in the back near the door is another place (but not as good as the forward seat). Don't sit underneath the wings, this will get you the worst reception.

Cheers.

Costyn van Dongen - http://www.flylikebrick.com/ - World Wide Wingsuit News

[KrisFlyZ 0]

For £40 more you can get an Edge 205. It is a fine unit and there are no hassles with satelite acquisition. I own a Foretrex 201 and edge 205....can't even be bothered to use the 201. However if you are patient it does work. I have one for sale .

Kris.

[Waldschrat 0]

We sat both underneath the wings, because tandems and a few other wingsuits sat in the plane.


@Krisflyz
Did this Edge 205 also works without interruption in 206 or 182?

[Kookaburra 0]

Just what I had hoped someone would design: real time feedback of glide angle and analysis of the flight right after landing... so, yes yes yes, I'm interested. I have been using the GPS 10 and Pocket Dell Axim X 51 v (not a Palm though), but analysis with Paralog gave me some weird results, then, reading the Axim specs, I found out that the maximum operating altitude is 10,000 ft! Would another 3 or 4 grand really prevent it from working?
Thanks so much for sharing!

[crwper 7]

Quote

Waldschrat: Your are not in trouble with the pressurization with this plugs in your ears?

I haven't noticed anything. Maybe the small hole down the the middle of the earpiece (for sound) is also allowing pressure to equalize.

Quote

I hope you add my garmin foretrex in the next time.

The problem is the data cable. Initially I experimented with a Garmin Geko using serial cables to attach to the universal connector of the Palm. I think maybe 20% of the time the connector actually stayed in place through gearing up. Most of the time, though, I'd be walking to the door and the cable would come undone.

Quote

bofh: A guy (Björn Alving) at my dropzone (FKCG, Sweden) has developed his own hardware for this.

I thought of doing it that way for a couple of years, but I'm not that good with electronics so it looked like the project would never see the light of day. Then one day I realized I could turn it into a programming project. My hat goes off to anyone who can do this in hardware only.

Quote

Costyn: Does the Palm have an USB interface? in that case it might be possible to hook it up to a GPS with serial interface (like ForeTrex models), and output realtime data in NMEA form?

I'm currently using NMEA data through a Bluetooth serial connection, so it wouldn't be that hard to go through the Palm's universal connector as another option. I had dismissed this option because of difficulties with early prototypes, but I will add it back in, since interest seems to there.

It looks like the weather here is taking a turn for the worse, so maybe I'll have more time than I expected to work on things this weekend.

Michael

[yuri_base 1]

Michael,

Can you regenerate the sound from the recorded data after the flight and post an mp3 file and glide ratio graph here? It would be interesting to hear what you hear.

How do you handle the noisyness of glide ratio data? With Vista C, I found that at least 5s running average is needed to provide any meaningful data.

Also, when using your setup, wind should be taken into account. Flying into a strong wind favors faster forward speed for better ground glide ratio and that is not necessarily better L/D. Flying approximately perpendicular to wind will greatly reduce the difference between GR and L/D.

Yuri

Android+Wear/iOS/Windows apps:
L/D Vario, Smart Altimeter, Rockdrop Pro, Wingsuit FAP
iOS only: L/D Magic
Windows only: WS Studio

[KrisFlyZ 0]

I cannot answer that question decause I have not jumped either airplane but the 305(same as 205 with barometric altimeter) worked very well in a Beaver.

Back on topic.... Please post an MP3 and GPS data like Yuri suggested. It would be very interesting to see as well as hear from jumpers that use this system how it is working for them. If I had a GPS 10 I would definately give this a try but I own too many GPSes and will not buy another .


Kris.

[crwper 7]

Quote

KrisFlyZ: ...but I own too many GPSes and will not buy another.

I will definitely be adding support this weekend for non-wireless GPS units and those which do not transmit the PGRMV (Garmin proprietary 3-D velocity) sentence. You'll still have to figure out cables, but it will lower the barrier a bit.

Quote

yuri_base: Also, when using your setup, wind should be taken into account.

I was thinking the same. I have a couple of thoughts on this... One is that the main piece of information is relative change, so you can see if you are improving your glide angle with a change in body position. In that case, wind effects wouldn't really matter too much.

However, since the sound is scaled to cover a range of glide angles, it is possible that the wind could put you outside that range. One possible solution is to have an option to subtract some bias from the horizontal velocity. This would be an assumed wind velocity at altitude. However, I'm still unsure if this would be helpful or just confusing.

The options are limited only by the data we have to work with. Everything else is just a programming problem. I'd like to start simple, but would love to hear suggestions from anyone with experience interpreting freefall GPS data.

Quote

How do you handle the noisyness of glide ratio data? With Vista C, I found that at least 5s running average is needed to provide any meaningful data.

I've been using the PGRMV sentence (mentioned above). This provides 3-D velocity data. Up to this point I've assumed the GPS used phase drift information to calculate this (since they specify an accuracy of 0.1 knots), but on further reasearch I think they may just be smoothing the position data. In any case, for units which do not output the sentence I'll need to calculate distances (using the haversine formula) and do some smoothing myself.

I'm not too concerned about window size. Even a 5-second running delay in feedback would give a lot more info than is currently available. Any suggestions on the smoothing function?

Quote

Can you regenerate the sound from the recorded data after the flight and post an mp3 file and glide ratio graph here? It would be interesting to hear what you hear.

I'm sure I can come up with something.

Michael

[LouDiamond 1]

For those who are interested in this type of feature and would like to test something similar out, I suggest you contact Klaus at Paralog as he has a working prototype that we have been playing around with as well.If enough people are interested it will no doubt speed up the development.

The software works with Bluetooth equipped GPS using the NMEA protocol without the Garmin specific sentences.The software itself runs on any Java ME equipped device which is basically any Windows PocketPC, Palm and most of the current mobile phones.Right now it displays realtime data, the audible feedback is not only for glide, but also for horizontal and vertical speed. It will more than likely have a user definable reference/neutral point so it can display changes to a preset value and the user can choose what information is made audible. Of course, all of the collected data will be able to be downloaded from the PDA to Paralog on a computer for post flight review and comparison.

"It's just skydiving..additional drama is not required"
Some people dream about flying, I live my dream
SKYMONKEY PUBLISHING

[Kookaburra 0]

Go ahead, Michael, we are still interested in your project!

[crwper 7]

Quote

yuri_base: How do you handle the noisyness of glide ratio data? With Vista C, I found that at least 5s running average is needed to provide any meaningful data.

I've been doing some experimenting this weekend looking for an alternative to the proprietary 3-D velocity sentence. I can get horizontal speed in the NMEA format from, for example, the GPVTG sentence. However, I'm still stick with taking finite differences of altitude to get vertical speed. The result of this method is nowhere near as accurate as using velocity data directly from the GPS.

I've done more research into velocity determination in GPS units, and came up with this. The velocity data calculated by the GPS and returned, for example, in the proprietary Garmin sentence I've been using, is actually calculated from phase shifts in the carrier. This is considerably more accurate/stable than velocities derived from finite differencing.

Other than the Garmin proprietary sentence, I can't find any NMEA sentence which contains a 3-D velocity. However, there is a single record in the Garmin Binary format which contains both 3-D position and velocity data, and should be output by any Garmin GPS unit. I am considering the switch to Garmin Binary format for this reason.

My question to everyone who might be interested in using this software is: What make/model is your GPS?

Michael

[KrisFlyZ 0]

Quote

My question to everyone who might be interested in using this software is: What make/model is your GPS?

Michael

Even if it does not have Bluetooth? I use an Edge 205 or an Edge 305.

Kris.

[crwper 7]

Quote

KrisFlyZ: Even if it does not have Bluetooth?

Even if it doesn't have Bluetooth. Reading from the cradle connector on the Palm is almost identical to reading from Bluetooth, so I've now got that part implemented. Of course, you'll have to have the appropriate cables, but that's cheaper than a new GPS at least.

Michael

[Kookaburra 0]

Quote

My question to everyone who might be interested in using this software is: What make/model is your GPS?


GPS 10 and pocket PC DEll Axim X51v
JJ

• Quote

[Costyn 0]

Quote

My question to everyone who might be interested in using this software is: What make/model is your GPS?

Garmin ForeTrex 201

Perhaps if more people are going to be using it and you want to make it open source, we can set up a SourceForge page for it?

Cheers,

Costyn.

Costyn van Dongen - http://www.flylikebrick.com/ - World Wide Wingsuit News

[Gadget 0]

My question to everyone who might be interested in using this software is: What make/model is your GPS?


Garmin foretrex 201

If people from Poland are called Poles, why aren't people from Holland called Holes???
My logbook

[cpoxon 0]

Quote

However, I'm still stick with taking finite differences of altitude to get vertical speed.

Is this accurate enough for the intended use her? From GPS Altitude Readout > How Accurate (which is admittedly quite old), it says:

Quote

Altitude error is always considerably worse than the horizontal (position error).

So with the altitude error compounding the horizontal, plus a lag from averaging, are you getting useful data or are you flying reacting to data that was wrong 5 seconds ago even before it was smoothed out?

Skydiving Fatalities - Cease not to learn 'til thou cease to live

[crwper 7]

Actually, as it turns out I don't need to do the averaging at all. Here's why:

If I use finite differences on position then, yes, the results are very inaccurate. The positions themselves carry a bit of error (the vertical about twice the horizontal), and finite differences compound the problem. To get reasonable glide angles, you need to average over a few seconds.

However, I have been able to use a proprietary Garmin 3-D velocity sentence. This velocity is actually calculated in the GPS using the drift of carrier phase (i.e. Doppler shift), and is accurate to about 0.1 knots. At freefall speeds, this is plenty accurate for my purposes.

The problem is, most people don't jump with the GPS 10, which as near as I can tell is the only one using that sentence. So, I'm currently porting my code over to the Garmin format instead of NMEA. In that format, the GPS actually sends a single packet of data containing time, 3-D position, and 3-D velocity. The best part is, every Garmin GPS made today is capable of sending that packet.

There will still be a bit of lag. There's about 1 second lag just because the GPS doesn't send data that often, and there is an unknown amount of smoothing done in the GPS to calculate velocity (velocities are calculated in a "black box"). However, the tests I've been able to do so far seem to indicate that there is no more than a couple of seconds' lag in the data. So you just have to hold a position for a couple of seconds, and you'll know if that's a better flying position.

Michael

[crwper 7]

Quote

Perhaps if more people are going to be using it and you want to make it open source, we can set up a SourceForge page for it?

I think Sourceforge might be overkill for this one, but I'm open to the idea. Will send a PM. I should have the current porting work (to Garmin instead of NMEA) done by the end of the week, and will be posting the code I've got, as well as PRC's.

Michael

[yuri_base 1]

Sweet! The next step would be to hack firmware and make it sample velocity 10 times a second.

Here's an interesting site:
http://transportation.hackaday.com/category/gps/

and particularly this idea about using radiosondes as cheap Doppler shift GPS trackers. More info here:
http://www.digitaldawgpound.org/nick84/post=130
http://www.digitaldawgpound.org/nick84/post=133

Android+Wear/iOS/Windows apps:
L/D Vario, Smart Altimeter, Rockdrop Pro, Wingsuit FAP
iOS only: L/D Magic
Windows only: WS Studio