gps for chopper

[kelly steed]

need to take photos off a constrution,the unit being construted is vauled at 14 mil . client needs four shots per week from four differnt locations,i would like to take shots using same location each time.time for build is six weeks,any way i can use gps to keep location of shots the same each week.

[GGoodrum]

Check here: www.u-nav.com

[spork]

...any way i can use gps to keep location of shots the same each week.

Yes, but this can be very tricky. Generally, unaided GPS is only accurate to about 10 meters at best. Differential GPS can get you down to the realm of about 1 meter (this requires a GPS base-station at a fixed location and another GPS RX on the heli). GPS in carrier-Phase mode can get you down to about 2 cm's (this is non-trivial unfortunately).

Depending on how far you plan to be from the building perhaps straight up GPS at 10 meter accuracy will be sufficient for your purposes. However, the GPS just spits out positions (typically NMEA ASCII strings). In order to make the heli go to a location you have to read those strings (easy) and incorporate them into the control-loop of the heli (not easy).

Some of the best GPS RX's for you application are made right up there in the Great White North (by Novatel). But if you're thinking about carrier-phase stuff, the prices start at around $10K. Unfortunately, Novatel won't be able to tell you much about integrating GPS into the heli's flight controls.

Ga Tech ran an "aerial robotic vehicle" competition for a number of years. The typical vehicle was an R/C heli with GPS (and other) guidance and control sensors. Check out this link:

http://avdil.gtri.gatech.edu/AUVS/IARCLaunchPoint.html

One of the toughest parts to this is sensing your attitude. To sense attitude with any accuracy requires some fancy footwork if it's to be done with GPS (the Trimble "Tans Vector" is a good place to start).

[MarkWebber]

Check here: www.u-nav.com

I didn't see anything there that was for heil's. Maybe I missed it but, it all look like fixed wing setups.
If it's for your own reasons and not at the request of the contractor, I would go too overboard spending on GPS. The contractor I've been shooting for doing the same 4 side shots is happy as long as it's generally close to the previous views.
I know I've seen less expensive systems but can't remember from where. The were working in congunction with video overlay systems.

[LoopBaCK]

Have you looked at the Eagle Tree Systems (http://www.eagletreesystems.com/UAV/uav.html). They have a GPS along with mony other options. I do not know if a video overlay is available (yet) but they definitely offer a wireless telemetry downlink.

[Menno]

Here are some more options:

GPS video overlay options:
http://www.dpcav.com/inspire/inspire.htm

non GPS video overlay options:
http://www.act-labs.com/black-box/
http://www.dpcav.com/vizion/vizion.htm
http://www.kapelec.com/altiose1.htm


I would like to see a hands on review if you decide to try one of these :glasses2:

Menno

[askman]

I don't think you need GPS for this job, but if you must, OSD (I use intuitive circuit OSD) is nice to use. otherwise, eagle tree is another option for ground based reading if you don't want to use OSD.

u-nav does not sell heli specific like carvec system. still, I think it is usable with something like spartan-rc ap2000i in heli.

[MarkWebber]

I think it is usable with something like spartan-rc ap2000i in heli.

That might be interesting. :idea:

[groundparrot]

:arggg: Just setup a live video feed and match the view to the picture you took last time.
I have used this method to take hourly construction photos that were later animated.
Keep it simple. :roll:

[Angelos]

Waypoint navigation with fixed wing is easy because the aircraft is always in motion. Thus you can work out the heading from the GPS data which is how the u-nav works.

To keep a helicopter in stationary hover you need to apply the correct corrections to the swashplate. To do this the system needs to know the heading of the aircraft and GPS can’t be used here as it does not provide heading when stationary. Magnetic compass (electronic one of course) is the solution but to avoid excessive errors in measurements you need to know the pitch and roll of the aircraft. This then requires an AHRS to calculate the attitude of the aircraft.

Recently I found a paper that describes a method about calculating the heading using a 3 axis magnetometer without the need of knowing the pitch and roll angles. I am still sceptical if this will work because none of the well known compass sensor manufacturers suggests this method. I need to try it and see for myself. If this works is will be possible to add a reasonably priced GPS position hold it the AP-2000i with IR sensor.

-Angelos

[GGoodrum]

VERY cool, Angelos. I look forward to your test results. :)

[groundparrot]

Angelos,

Wouldn't the HH gyro give you the reference you need?

[woodturner]

I second groundparrot's comment. I agree GPS and other NAV systems are really cool, I work in this field. However the simple approach may be the best.

You can implement something simular to what the Hubble Telescope uses..

You will need a video downlink of some type. Go up and take your initial shots, you will need to include back gound objects you can reference next time. You'll also need to mark the position you are flying from.

Print the images from your first shoot & take them on-site.

Now using the video downlink line up those background objects & bingo your bird is in the same location as last time.

Simple, effective & low tech.

Good luck which ever road you take..

Happy flyin'

Frank

[Angelos]

Wouldn't the HH gyro give you the reference you need?


Assuming the helicopter will go from exact forward flight into stationary hover a method similar to HH gyro can learn the last GPS heading and maintain it. However this will be for short periods of time only; the gyro will eventually drift.

The question is what happens to the helicopter when the actual heading does not match the one computed by the HH gyro. Or if the pilot does not switch on it immediately after forward flight but after a sideways of backwards move. The entire thing gets really screwed up and the system will almost certainly loose control of the aircraft.

-Angelos

[catfight]

How about another idea (untried)? Place a battery powered laser pointed straight up - and have an onboard sensor that tells (beeps, flashes, etc.) you when you are over that spot (no elevation data)? Only usable of course if you can reposition the laser each time- which you could do with a still photo of the spot usually:D Of course with 3 lasers you could (gets ridiculous here,,have elevation data).

[chopper jockey]

all you have to do is tether the airframe with a strong fishing line at the required length. climb untill the line is tight. Mark the ground and return to the same spot later with the same length. :wink:

[Angelos]

From what I’ve see most people don’t need GPS to acquire the same shutting position every time. Instead they want the helicopter to maintain it’s position so they can look at the downlink and frame the shots solo.

-Angelos

[GGoodrum]

I wholeheartedly agree, Angelos, the bigger market is to have a system that will hold its position, wherever it is. The only one that can do that now, that I'm aware of, is the $10k Carvec system.

[Yukonho]

For your immediate need, (depending on altitude required) why not use a mast? Easy enough to put the mast in the same place each time...
I know this isn't what you are asking for, just another way to skin a cat.
Colin