Hovering to within 50ft

[mickjoebill]

Looking for help with a concept.

I have a long term photographic detial that requres hovering at 1500 to 2000 feet for 20 minutes. Stable non drifting hover essential. Aircraft nose has to remain within an approximatly 50ft imaginary cube. We may be able to get away with 100ft cube if we go higher.
Roll and yaw movements will be corrected by camera gyro.
Orientation of aircrfat can be in any direction.
SO aircraft can be wobbling quite a bit it has to stay within the box.

Ideal aircraft choice for camera is AS355.

In the long term the Pilot aid would need to be derigable and installed in local aircfaft worldwide. We have experience in packaging kit and making it derigable including putting payloads on exterior of aircraft.

Short term we could play with doppler equipped helicopters.

Looking for ideas on how to do this!
Daytime flight.
Flights can be preplanned with ground assistance maybe something like home made papi lights would help???


Is some form of portable derigable doppler display possible? if we think it feasible for pilot to fly referring to this. (assuming doppler cant easily be integrated into an autopilot equiped AS355 as in AS555 version)

Larger helicopter with doppler autopilot could be considered as last resort but we will be over built up areas so noise is a big issue as is fitting our camera.

Mickjoebill

[Art of flight]

Good luck with this!

I've flown 355 in the corporate and police role with and without AP and it's a very stable camera ship, just not aware of any modification that would assist in maintaining those parameters. As for doppler aids, i used those in Gazelle and Lynx and from 1500' I don't believe the pilot would have sufficient rate of change information to maintain the 50' box by using those aids alone.

IMHO you're probably best off just employing an experienced filming/police/news pilot who is used to maintaining that level of accuracy for 20 minutes or more and hoping the weather is good.

Currently fly the EC135T1 and T2+ in the police role (lots of high hovering day and night) and would say it is almost impossible to maintain a hover within 50' at 1500' for 20 minutes (not for the lack of trying!). That aside still prefer the AS355 for camera work.

regards

[tecpilot]

Forget doppler in 1500-2000ft for hovering within 50ft range. The aberration of doppler in your time limit and in this altitude is to much. May be a special DGPS unit with near based ground station will have the needed accuracy. 15ft in 1000ft i believe as published but only for a fixed measurement from a fixed measuring point with digital indication, but not for measuring in an any time at least small moving helicopter.

Is the SAGEM 85T31 AP now with hovering mode??? :=

[HVHmt]

I agree with techpilot, it can be done but you will need the Differential GPS. as typical aviation units generally require a velocity component.

Check with Agricutural pilots, or a local survey company. They use this equipment on a daily basis.

[SASless]

Did something similar a few years ago....we tricked to usihg the installed FLIR and pointed it vertically...whalah!

The hard part was maintaining height as we to do 500 foot increments to 3500 feet to satisfy the customer.

[Jackboot]

Managed an out of wind hover at 1200ft for 20 minutes in my R44 to get the Vulcan take off at Bruntingthorpe. I am a low time PPLH.

The only problem was yawing to the right to allow the video cameraman to pan. Wind was IRO 10kts - enough to 'snatch' as we turned. The stills from the rear seat guy were fantastic though.

Hard work but do-able on a good day if you dont need to turn in the hover.

Jack

[Gomer Pylot]

It Europe it may be hard, but in the US, with WAAS, the GPS altitude readout should keep you within limits. WAAS accuracy should be less than 50' almost all the time. The GPS will need to be in the pilot's field of view, of course, not mounted in the console. Something like a Garmin 330 or 430 mounted on the glareshield should work. On the other side of the world, I have no ideas.

[[email protected]]

Use a helo with a Rad Alt hold and a heading hold - that's two less variables to deal with. Then it's up to pilot skill since the spread of the doppler beams from 1500 feet will be too large to give the required accuracy.
Inertial nav fed by RLGs perhaps? I don't think the GPS height is accurate enough for your given parameters.

[tecpilot]

Crab, old boy stay in low alt! Which radar altimeter will have a accuracy of 50ft in 2000ft? The most i know have just at least an indication in such altitudes in 50ft steps, plus technical tolerance. And your heading hold is nuts. In 20 minutes i fly easy 5-10 miles groundtrack allways horizontal any time with fixed heading.

And a DGPS will give a much more correct altitude than a normal GPS.

:) :) :)

[rotarycat]

Sounds like something I have had to do a couple of times at motorbike races to relay the signal from the cameras on the bikes.
Most pilots should be able to maintain the helo within those parameters if the vis is reasonable. I was using a B206 and was at 2000'.

[Nubian]

I think mickjoebill was looking for a CUBE (50x50x50ft) not if a pilot can maintain a high hover at 1500-2000 ft. within 50ft. tolerance.
When you get to that sort of altitude, you can sit in what you think is a perfect hover (maintaining altitude) but still drift for/aft, sideways those feet without you notice it, and be well outside of the 50ft lat.long before you adjust for it. Your periferal references at such height is just not sufficient.

[Buitenzorg]

mickjoebill,

I used to work in the seismic industry and I’ve done a bit of bird towing (best 3 weeks of my flying career thus far). There are several short-range navigation systems available that will give you the required precision; almost all are being/have been phased out and replaced by some (often custom-made) form of differential GPS.

During the bird towing I was required to fly within 25m laterally of preplanned lines, at about 60 kts or 40 kts depending on the type of bird, as low as possible without wiping out the bird against a tree; after a bit of practice it was rare to be more than 5m off.

A display like a large (about 4” X 6”) PDA was mounted to the right of the instrument panel, showing the survey lines and the aircraft’s position relative to them. The position data points were captured with a precision of well under 1m in the x, y and z axes. The actual navigation system was differential GPS.

I don’t know of any off-the-shelf navigation system that will allow you to do the job you want but based on my experience I’d say one can be easily created. It will involve a differential GPS setup (a GPS base station on the ground sending correction signals via radio link to a “rover” GPS receiver in the aircraft) and a display for the pilot showing corrections to be made in X (forward-back), Y(left-right) and Z(up-down) axes. As your mission is photography the flights will be in day VFR conditions with good visibility and ground reference, so aircraft control will not be a factor. Since your camera mount will allow photography regardless of the aircraft heading you can just hover into the wind; with a known hover heading the position information can be easily converted from north-south, east-west, to the X and Y axes mentioned above.

Any land surveyor who is computer literate (and today that should be any land surveyor) can whip up a program to give you the above display in half an hour or so, and they are also used to working with mobile DGPS systems. I’d suggest getting into contact with a land survey company (come to think of it, a company called Fugro do a lot of aerial surveying worldwide, I’m sure they could get you pointing in the right direction) or the exploration department of a mining company and explaining what precisely you’re after.

Edit to add:

I've just been on the Trimble website (www.trimble.com) and it would seem they have handheld receivers/computers which could fit your bill even in basic (non-differential) mode with the right antennas. Try giving their UK representative(s) a call.

[NickLappos]

Buitenzorg has it right. No ppruner on earth has held a helo within 50 feet of reference while at 1000 feet, let alone higher, the visual cues are way too poor for that accuracy.

There are many survey systems that have a small DGPS transmitter that you set as a reference at a construction site, and then read the corrected GPS position within an inch or less.

Displayed on a crosshair X-Y plot on a laptop or handheld and you can easily hold reference if the winds stay steady.

Here are some examples:
http://www.trimble.com/geobeacon.shtml

http://www.trimble.com/geoxh.shtml

[helipedro]

Send me a pm maybe I can find some info. A few years ago I met a guy that used laser beam. I believe hi still has the system stored somewhere.It really worked

[sunnywa]

What I love about this forum is that, apart from the occasional verbal sparring match that doesn't seem to go anywhere, is that I learn a lot from it. It is very good to learn how there are DGPS systems that could do this. :D

My only question though is why do you have such fine tolerances mickjobill? as it seems to me to be some techie saying I want this exact thing when a bit of latitude would not change the angles very much and give pretty much the same picture.

[albatross]

Have to disagree with you Nick.

While too long and involved to help Mickjoebill

We used to do "laser survey" in which you hovered over a laser aligned verticaly. The laser shone on a screen under the helicopter and showed up as a dot on a small TV screen.

You hovered low over the laser till you saw the laser "Dot" then climbed verticaly keeping the dot in the screen - you had about 2 ft to play with.

Basically you were acting as a stadia rod for a surveyor on the ground.

If you lost the dot you had to descend back and pick it up over the laser again.

It was done initialy using a Gazelle or an Allouette with a system of mirrors in a box on the nose to see the laser beam. We did it later with AS350 and H500s using the TV setup.

It took about 10 -15 hours and a lot of swearing to learn how to do it. Once you got the nack it was quite easy to climb to 3000 -or 4000 and stop for the surveyor to get his shot - altitude was easy to control and you would be within +- 2 feet of position.

I recall getting to 7000 ft plus AGL for 1 shot but the average required was probably 3500.

Going up vertically and stopping were not too bad - descending down the beam was very hard and could also lead to a great demonstration of settling with power.

This was in the 70 and 80s - GPS killed it - all the cowlings and stuff are probably moldering in the back of a hangar somewhere in Quebec and the skills are long rusty.
Sure was fun at the time.

[NickLappos]

albatross,
I do agree, a laser system would surely work well.

[helicfii]

I say find a nice mountain that doesn't move and put the camera on it instead :8

Seriously, though- all you need is two guys, each with a rifle scope on a tripod. One on his back looking up at the helicopter, and one in the best lateral position you can find (is there a building in the vicinity? If not, then he has to position as far away as practical and utilize a diagonal view for altitude).

Give each one a radio.

The guy on his back is call sign "position", the guy in the lateral position is call sign "altitude". They could keep you in a 50 ft box with a little bit of practice. Just have them give you vectors.

Ex: "Position- correct left" "or Altitude- correct down"

Don't drink coffee beforehand, and try to stay relaxed. :ok:

[Matthew Parsons]

The problem with the laser system is it doesn't give you altitude within 50'. If you modified it with a second laser that the helicopter intercepts during the climb, then it is feasible that you could maintain the cube. Once the pilot has practiced enough, the most difficult part of this would be to get the position laser pointing perfectly vertical, and getting the altitude laser intersecting the position laser at the desired altitude (within 50'?). Not impossible, but the altitude adds a degree of difficulty.

Using two scopes is theoretically possible, but would be very difficult. The terminology could become cumbersome. Getting into both scopes initially may be difficult. Also, 50' at 2000' distance is about 1.5 degrees of arc. To give the pilot any useful accuracy, you would need quite a bit of magnification. Then the problem of getting into the scope becomes worse. Also, you can use that to maintain the position, but how do you first set up the tripods with any accuracy?

If the DGPS can't work, and the area is relatively flat, I think the best option would be to install three laser range finders on the helicopter, and process the distances to give you position and altitude errors graphically (i.e. cross hairs on a PDA). Fairly inexpensive range finders can give you meter accuracy up to about 2.5km.

Other ideas are possible, such as determining the distance from ground stations using RF or ultrasonics (not sure if that would work at 2000'), however those would require quite a bit of development, and then calibration would be difficult.

There is a low tech solution that may work out. Build two large square frames with a bright light behind each of them. Do the trigonometry and position the lights and frames such that the only way the pilot can see both lights within the respective frame is when he is in the 50' cube.

Based on some calculations on the back of an envelope, if the pilot can look 30 degrees below the horizontal, and can scan between -45 and +45 degrees of the helicopters nose, you need to place the lights on the ground ~3400' from beneath the helicopter, and set the 15" square frames 50' above ground and 85' in front of the lights. All the pilot has to do is keep the lights inside the 15" squares that are about 3/4 of a mile away (line of sight).

Okay, maybe it won't work out, but hopefully those numbers will demonstrate how difficult the desired accuracy actually is.

DGPS or a tall stick in the ground seem like the best answers.

Matthew.

[Angels 60]

Deleted by SP: irrelevant rubbish

[albatross]

To further beat a long dead horse:

Actually the laser we used was a battery powered 10 amp gyro stabilized to vertical unit on a tripod so straight up was straight up. It was best to check with a small level on the lens on a cold day as it sometimes froze off level and straight up wasn't - the surveyor would sometimes do his thing and then ask you to climb or descend on the beam to confirm that you were indeed going straight up or down.

If you set up in a swamp and someone walked near the laser you would actualy see the beam start to jiggle and a quick call on the radio to have the person stop moving around was needed.

When the surveyor ( located miles away saw us appear over the hill he would tell us to stop climbing and we would hold postion on the laser Dot in the TV screen and altitute with the altimeter.) Holding altitude once hover was achieved was not a problem - at least the surveyors never complained.

Once the shot was done we would descend - not looking down at the laser of course and pick it all up then move to the next turning point and do it all over again.

Anyhoo - all done a long time ago in a Power project far far away.

[Jackboot]

Hi Silsoe,

We had plenty of time to experiment try various positions - the thing snagged during backtrack if you remember. I had a still cameraman and a video cameraman both giving me various up, down, forward, back, left right instructions right up to the wire.

we were also out of wind.

My view is that a low time 44 pilot should make a reasonable job of staying within the parameters discussed, especially if the wind is on your nose.

Obviously, a more sophisticated and expensive machine will be preferable.

How much profit do you want to make in completing the task described though?

I have already seen some of the images from Bruntingthorpe. Pin sharp and just awesome.

Jack

[mickjoebill]

My only question though is why do you have such fine tolerances mickjobill? as it seems to me to be some techie saying I want this exact thing when a bit of latitude would not change the angles very much and give pretty much the same picture.
Project involves very high resolution images, can't say more than that:)

Thanks for the feedback, great stuff!
Reason Im asking is that we need a reliable solution that will work every time.
Ive flown with the best aerial filming pilots in the world and they do not get called upon to fly this kind of profle, it is more difficult than it appears especially for 20 minutes. Some locations would be easier than others in respect to visual clues.


Should we consider two pilots for both saftey (keeping lookout) and performance?

At 2000 ft the box can be bigger. 100x100x100ft. This is probably the practical altitude limit.

So doppler and radar alt are non runners.
I'll follow up the laser system although it may be far from practical to install.
DGPS seems the most practical to install so Ill give that a go first.

The most versatile sytem would enable us to select the cube once we are airborn and over the target, so we can fine tune the position before we commence the actual recording. The position of the cube itself needs to be fine tuned to within 100 ft or so, depending on subject.


Since there is mixed opinion as to how easy it is to fly "the cube", it sounds like it would be a fun compeition to run at a fly in!


Thanks Mathew, Albatross and Nick Helicfii, Buitenzorg for the feedback

Mickjoebill

[Gomer Pylot]

The laser can give you altitude to within millimeters. Scientists have used lasers to determine the distance to the moon, and most common rangefinders use lasers these days. This is one the most accurate ways known to determine distance. Determining the altitude isn't the hard part, displaying it so the pilot can maintain it is harder, and the hardest part of all is staying at the correct altitude, and over the correct point. A 4-axis flight director, coupled to a suitable electronic device, should be able to do it easily, but the cost won't be cheap. The US Coast Guard's helicopters can do this from low altitudes, but I don't know about the altitudes required by the OP. I think some sort of laser equipment would be the best way to go, if the funds are available, and it will require a rather large pile of them, I suspect.

[[email protected]]

Gomer - as has been pointed out to me - what you can do in a SAR helicopter with a 4 axis FD at lowish levels (my rad alt hold works up to 999' in the hover and quite accurately too) won't cut the mustard in terms of accuracy at 1500 to 2000'.

Having said that, my Bar Alt hold will maintain a pretty good height in the hover at 1500' (we use it for FLIR searches) but it does have accelerometers helping to fine tune the barometric information.

Equally the heading hold is good for +/- 2 degrees in the hover - the only problem would be the ground position - back to lasers I guess (or really good trimming:))

[JerryG]

What a fascinating exchange!

I'd rule out GPS. The last thing you want in an accurate hover like that is to be moving your head (ear canal fluids) or changing your eye focus from 2,000' to 2' to read the screen and back out again. In a two year period of working on trial drops of the parachutes for the NASA Cassini Huygens project I concluded that your head had to be either inside the cockpit or out, but not both. In that particular case we had the luxury of being able to do it at 5kts forward ground speed (not applicable in this case) so my head stayed inside on the GPS during the approach and my observer kept his eyes outside for safety and confirmation.

I'd go with an additional person in the cockpit calling out accurate heights from a rad alt and then concentrate visually on the drift within the remaining two axes by looking ninety degrees out to the right. The double light system suggested would be perfect for fore and aft movement, but I've done it a few times just by lining up a near object and a far object, it works OK. You'd certainly achieve considerably better than 100' accuracy without needing the lights.

For left and right drift you are simply lining up the bottom of your window (or a chino line you've put there) with something on the ground and keeping it at the same position in the window. You'll soon notice if it moves.

The best tip in the exchange above is to settle down. I'd add to that to consciously control your breathing and most important of all....don't move your head under any circumstances. (This proved quite hard on the day a car plummeted past me, having rolled off a cliff-top whilst I was doing a close proximity cliff rescue. The next twenty minutes were largely occupied with looking forward to finding out "what the hell was that flash of red in my peripherals"?!)

JerryG

[rojread]

Mickjoebill

Came a bit late to this thread. Suggest you contact Prof Albrecht Grimm at IGI GmbH, Kreutztal, Germany. He is the whizz-kid kid on DGPS aircraft positioning systems and loves new problems.

[email protected] or igi-ccns@t-online

IGI pioneered Differential GPS based survey aircraft positioning systems and x, y and z positioning sub 10' is their forte. I've flown hundreds of hours with his toys and never missed a pic yet.

Good luck

[McBad]

"Larger helicopter with doppler autopilot could be considered as last resort but we will be over built up areas so noise is a big issue as is fitting our camera."

Just curious, in this era of Big Brother and one cctv camera to every 14 members of the population, as to WHY you want to hover over built up areas to take photographs with that level of precision.

Cheers,

M

[mickjoebill]

"Larger helicopter with doppler autopilot could be considered as last resort but we will be over built up areas so noise is a big issue as is fitting our camera."

Just curious, in this era of Big Brother and one cctv camera to every 14 members of the population, as to WHY you want to hover over built up areas to take photographs with that level of precision.


I may be being over sensative about the noise issue, but the kit would have to travel to various countries so AS355s would be aircraft of choice due to availability of aircaft and brackets and we could take our own handling pilot. I think using two pilots is a no brainer.


Why do it? There is money in it if we can make it work :) We wont be doing council housing estates:ok:

Mickjoebill

[tomstheword]

problem solved, all you would need then is an experienced driver (a precision long liner) and a bag of sand at the bottom that stays on the ground to keep the line taught. The angle the line is on would give you the lateral drift and the sand the height, you could even put two bags on the bottom spaced 50' apart so one would be off the ground. Do I get a commission.