SMALL - espritmodel.com SMALL - Telemetry SMALL - Radio
Reply
Thread Tools
Old Dec 25, 2008, 09:37 AM
Registered User
United Kingdom, Bath
Joined Apr 2007
193 Posts
Quote:
Originally Posted by mmormota
Good find, I don't know any better for an evaluation kit.

I proposed the magnetic sensor because the drift is inherent feature of the piezo gyro's and practically impossible to avoid it. The smm gyro's are better but more expensive, and hard to find smm sensor's in small quantities.
Good point. I think tilt compensated magnetometers deserve to be looked into, the gyro drift problem is (again) quite annoying... A 2 axis one is not quite all that's needed though, to do this properly you need a 3 axis magnetometer. Although I've found that it is possible with only 2 axis using an algorithm in a paper, which estimates the 3rd axis information by knowing the local magnetic inclination.

As with any magnetometer based solution, there is a new problem. If you live in a place with a lot of magnetic inclination, it will potentially limit how far you can pitch up/down. It will be great the lucky bastards living around the equator, not so great in Canada.

Here's a map of magnetic inclination of the world: http://www.ngdc.noaa.gov/geomag/WMM/data/wmm-I05.pdf

For example, in the UK, where I am, I=66 dgrs, meaning the for pitch > ~20dgrs the compass becomes useless even with compensation. What I think will be interesting is to use the magnetometer to correct the gyro when pitch is in a reasonable range, and depend on gyro alone when the magnetometer is in its useless range. If fact I'm going to order that SparkFun board to try is out when I get back to my place in the new year...

This will add another $25 to the cost of the HT... But if it works, at $75, it's still pretty cheap for an almost drift free tracker!
zitron is offline Find More Posts by zitron
Reply With Quote
Sign up now
to remove ads between posts
Old Dec 25, 2008, 09:46 AM
Registered User
United Kingdom, Bath
Joined Apr 2007
193 Posts
Quote:
Originally Posted by damian123
zitron, I liked that mission as I have done it myself I wish I would have a ht for the red bull race mission..
Heh, I've tried to do the redbull mission, I always mess up the touch and go part. I think for flight sims, I may try to amplify the turning motion. Right now it's 1:1, which I think is too much physical turning. I end up with a headache after flying for 15 min!

If anyone is interested, I have a modified SDK program that controls the camera in FSX (unfortunately won't work in any other MSFS) if you feed in serial COM data in plain text format.

Cheers,
-Z-
zitron is offline Find More Posts by zitron
Reply With Quote
Old Dec 25, 2008, 11:33 PM
UNREGISTERED USER :)
New York
Joined Apr 2007
773 Posts
yea amplifying the motion is a good idea.. all thought not to much either!

on the mission in that vid I made my touch and go without power but now I learned to do it will all the power all the time. Also I pull up for the vertical as soon I hit the gate.

If I would have a head tracker then sure I would love the modified SDK program but since I don't I don't mind using the original with the stop in the middle to make if perfect :P
damian123 is offline Find More Posts by damian123
Reply With Quote
Old Dec 27, 2008, 09:27 PM
Registered User
rcpilotjr's Avatar
Columbus, GA
Joined Feb 2008
110 Posts
What do i need to make the head tracker that zitron had for FSX?
rcpilotjr is offline Find More Posts by rcpilotjr
Reply With Quote
Old Dec 28, 2008, 08:07 AM
Registered User
United Kingdom, Bath
Joined Apr 2007
193 Posts
Quote:
Originally Posted by rcpilotjr
What do i need to make the head tracker that zitron had for FSX?

Hi,

As it is right now, you need:

A cheap RC heli gyro for ~$15
(from here for example: http://www.r2hobbies.com/products.ph...fffb0569a2bb5b)

A cheap 3 axis accelerometer board for ~$16
(from here http://www.sureelectronics.net/goods.php?id=176)

A cheap micro-controller board for ~$12
(such as the arduino http://moderndevice.com/RBBB_revB.shtml)

+ 20% for shipping and stuff
+ A method to program the above micro-controller
+ Code for the micro-controller*
+ Willingness to take apart a perfectly good gyro and figure out which are the important bits!
+ Some soldering/general electronic skills
+ Some LEDs, wires, resistors and pots
+ A software that interfaces with FSX*

*Freely available from me, but a their current state, may not work for you with out modification.

I must say again that I'm not developing this for flight sim use. The goggles I have are only 640x480, which is not enough resolution to read dials clearly. I'm only using FSX to test my head tracker.

Cheers,
-Z-
zitron is offline Find More Posts by zitron
Reply With Quote
Old Jan 27, 2009, 01:27 PM
Registered User
United Kingdom, Bath
Joined Apr 2007
193 Posts
update

I've managed to get the HT to interface with my RC tx, and built a pan-tilt platform to test it. :

My 3DOF head tracker controlling a Pan+tilt camera (1 min 17 sec)


I've put the thing on my RC car and driven it around a bit. Works quite well. I've added a digital high-pass filter to the gyro input, so drift is reduced a bit, but it's still there.

I'm getting some weird RF interference problem though. When the TX antenna is extended, the gyro starts to get a bias. The good thing is the bias is constant, so the high-pass filter removes it after ~20s. I'm using a lightly shielded audio cable to connect the HT to Tx, maybe I'll try a better cable...

-Z-
zitron is offline Find More Posts by zitron
Reply With Quote
Old Jan 27, 2009, 02:37 PM
Gravity is patient............
Joined Mar 2006
1,910 Posts
Quote:
Originally Posted by zitron
I've managed to get the HT to interface with my RC tx, and built a pan-tilt platform to test it. :

http://vimeo.com/2939439

I've put the thing on my RC car and driven it around a bit. Works quite well. I've added a digital high-pass filter to the gyro input, so drift is reduced a bit, but it's still there.

I'm getting some weird RF interference problem though. When the TX antenna is extended, the gyro starts to get a bias. The good thing is the bias is constant, so the high-pass filter removes it after ~20s. I'm using a lightly shielded audio cable to connect the HT to Tx, maybe I'll try a better cable...

-Z-
It's amazing that you have the technical know-how to do this. The head trackers available on the market are probably using the same cheap gyros you are. The difference is they have to mark up their costs to make a profit and stay in business. So if you stay at it and work out all of the kinks, you will probably end up with as good or better a unit then they have. Good luck with it! I am impressed.
billyd60 is offline Find More Posts by billyd60
Reply With Quote
Old Jan 29, 2009, 12:31 AM
Registered User
United Kingdom, Bath
Joined Apr 2007
193 Posts
Quote:
Originally Posted by Targetlocman
It's amazing that you have the technical know-how to do this. The head trackers available on the market are probably using the same cheap gyros you are. The difference is they have to mark up their costs to make a profit and stay in business. So if you stay at it and work out all of the kinks, you will probably end up with as good or better a unit then they have. Good luck with it! I am impressed.
Thanks, it's not that amazing . I bet there are probably several dozen people here in this forum that could have done this if they wanted to, and end up with a much better product and make some profit . I donno exactly what gyros the manufacturers uses, but If I was in the business of selling these, I would use expensive good quality gyros, simply because that way quality can be easily controlled, and you get less unhappy customers. I don't think they are making that much of a mark up...



----------

I've got a number of people asking me for the code, I've put the code along with some more technical explanations on the arduino forum here:
http://www.arduino.cc/cgi-bin/yabb2/...m=1231185714/0. If you have any coding specific questions, put it in the arduino forum. I'll try to keep this thread more FPV related.

Cheers,
-Z-
zitron is offline Find More Posts by zitron
Reply With Quote
Old Jan 29, 2009, 03:08 AM
Registered User
Daemon's Avatar
Lakewood, Colorado
Joined Aug 2002
28,932 Posts
What happens if you move it in two axes at once, and return on a different path?
Let's say the whole range of movement is represented by X-Y
coordinates where front and center is 0,0 and lower right is 10,-10
then try this path. Straight from 0,0 to 10,-10, then to 10,0 and back to 0,0.
Try that route a couple times in a row. I know it sounds a little
odd, but it's not uncommon to move/rotate head diagonally while flying.

ian
Daemon is online now Find More Posts by Daemon
RCG Plus Member
Old Jan 30, 2009, 08:58 PM
Registered User
Perth, Western Australia
Joined Dec 2005
327 Posts
Zirton, congrats on your work. How did you get the head tracker connected to your TX? Does it connect via the trainer port?
nick.sargeant is offline Find More Posts by nick.sargeant
Reply With Quote
Old Jan 31, 2009, 05:43 AM
Registered User
United Kingdom, Bath
Joined Apr 2007
193 Posts
Quote:
Originally Posted by Daemon
What happens if you move it in two axes at once, and return on a different path?
I *think* I know what you are talking about, and the answer is I'm not sure. I know the tilt will not be affected, since its from the accelerometer, but I'm not sure what will happen to pan. I do know there will be other issues when you look straight down or up so that your view vector coincide with gravity, at which point you loose the roll degree of freedom and trig functions may become unhappy...

I've put this project on hold for a while, been working on my DIY GPS OSD, tachometer sensor and generally messing around with my RC car, I will do some tests and see what happens.


Quote:
Zirton, congrats on your work. How did you get the head tracker connected to your TX? Does it connect via the trainer port?
Um.. yeah it's sort of like the a trainer port, in the sense that a wire in the form of a audio cable plugs into the TX, but it's completely different . You see, I've completely removed the original digital board from my TX, so now my TX is just a plastic shell running my own hardware and software. So I have many different options of getting my own HT to interface with my own TX. I could even use infrared or bluetooth if I wanted to, but right now the chip in the HT and the chip in the TX talk through simple wired serial communications at 38400 baud. The chip in the TX decodes the data from HT and appends it as channel 5&6.

It would take some extra effort to make my HT work with generic trainer ports, I currently have no plans of doing that, not least because I don't have a radio with a trainer port .


Cheers,
-Z-
zitron is offline Find More Posts by zitron
Reply With Quote
Old Feb 06, 2009, 07:34 PM
Registered User
United Kingdom, Bath
Joined Apr 2007
193 Posts
Quote:
Originally Posted by Daemon
What happens if you move it in two axes at once, and return on a different path?
Let's say the whole range of movement is represented by X-Y
coordinates where front and center is 0,0 and lower right is 10,-10
then try this path. Straight from 0,0 to 10,-10, then to 10,0 and back to 0,0.
Try that route a couple times in a row. I know it sounds a little
odd, but it's not uncommon to move/rotate head diagonally while flying.

ian
This is what happens if I look up and right simultaneously: pan will drift 5-20dgr, depending on how much I look up (see attached graph). Pitch has no problem whatsoever, as expected. The amount of drift is directly related to the pitch angle, the higher the pitch angle, the less sensitive the gyro is to the panning motion. If you look straight up or down, you can turn around all you want without the gyro reading anything, since the gyro axis is now reading roll and not yaw.

So as it is at the moment, the geometric drift becomes quite significant for pitch angles > +/-20 dgrs. Below that it's not too bad compared to other sources of drift.

There is one solution to this problem though, which is to simply multiply gyro sensitivity with a value that is a function of pitch (and roll) angle. This should compensate the drift for pitch angles upto 60-70 dgr ish. Above that there is not much that can be done apart from adding another gyro...

I will try to implement the geometric drift compensation (hey I need to trademark that term: GDC® ) when it gets warmer so I can think straight...

Cheers,
-Z-
zitron is offline Find More Posts by zitron
Last edited by zitron; Feb 06, 2009 at 07:40 PM.
Reply With Quote
Old Feb 06, 2009, 08:35 PM
Registered User
Daemon's Avatar
Lakewood, Colorado
Joined Aug 2002
28,932 Posts
Nice demonstration of the geometric drift problem inherent in gyros.
If you had a two axis gyro instead of gyro and two accelerometers
you'd find both the pitch and yaw off center at the end of that test.
I agree there should be some way to compensate for it by
looking at roll, but if roll is measured by accelerometer
then it becomes less accurate the further from level your pitch is.
I tried to work it out on paper once using 3 gyros
and it just seemed to be an endless feedback loop. Every
movement had a source of error that required a compensation
that had it's own source of error that had to be corrected itself with
another compensation.

ian
Daemon is online now Find More Posts by Daemon
RCG Plus Member
Old Feb 10, 2009, 04:55 PM
Registered User
United Kingdom, Bath
Joined Apr 2007
193 Posts
Quote:
Originally Posted by Daemon
Nice demonstration of the geometric drift problem inherent in gyros.
If you had a two axis gyro instead of gyro and two accelerometers
you'd find both the pitch and yaw off center at the end of that test.
I agree there should be some way to compensate for it by
looking at roll, but if roll is measured by accelerometer
then it becomes less accurate the further from level your pitch is.
I tried to work it out on paper once using 3 gyros
and it just seemed to be an endless feedback loop. Every
movement had a source of error that required a compensation
that had it's own source of error that had to be corrected itself with
another compensation.

ian
I'm using a 3 axis accelerometer, the pitch and roll angles are each calculated using two axes with a arctan2(x,y) function, and some filtering to get rid of noise, this way there is no degradation of angular resolution regardless what orientation the accelerometer is in. The accelerometer is great for the price. I wish the gyro was that easy...

If you use three gyros there is no reference you can use as a "trusted" orientation like a compass or accelerometer, the only reference you have is the starting orientation, that would be quite tough to do.
zitron is offline Find More Posts by zitron
Reply With Quote
Old Feb 17, 2009, 03:08 PM
Registered User
Joined May 2006
1 Posts
Modify the amplification on the gyro board

Hi zitron;

I have also been playing around with the E-Sky EK2-0704 gyro board, wanting to do head tracking. I'm interfacing it to a PIC 12F675.

Just stumbled over this tread.

A little tip I wanted to share. If you have not noticed, the final amplification on the gyro board is way to high for head tracking. Even wery moderate turning speeds applied to the gyro board peaks the output, and it flatlines.

I don't know if you have worked around this by tapping the amp before the final stage, but If you have not, you can broaden the dynamic range of the gyro board a lot by soldering a 0604 resister with a value somewhere around 7,5 - 8,5 K-Ohm on top of R15 on the board. I used three 22K's on top each other because that was what I had laying around.

This should make your tracker center better after a quick head-turn that before would have flatlined the output, and f up calcularions.

Please tell if this works, or you applied another solution?

Robert S.
Norway
la5xoa is offline Find More Posts by la5xoa
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Another very cheap head tracker device randall1959 FPV Talk 67 Aug 18, 2008 04:25 PM
Question Separate Radio for gyro Head-Tracker Crash9 FPV Talk 16 May 19, 2007 10:41 PM
Sold FS: Servos, Gyro, Brushless motor from Micro Heli rhetrick Aircraft - General - Radio Equipment (FS/W) 1 Jan 23, 2006 12:09 PM
DIY Heli Gyro mattweisz DIY Electronics 1 Jun 10, 2004 03:43 AM
cheap heli? SR_yoshi Electric Heli Talk 9 Jun 17, 2002 07:36 PM