Low cost 6DOF for this project

[uphiearl]

Just looking in here.
Has anyone done anything with the Sparkfun 6DOF Atomic yet ?

I have developed a 6DOF for < $35 using Wii nunchuck and Wii motion plus. Works better than the Sparkfun 6DOF !

I have been sidetracked on my quad and have been developing for flying wings.

My Parallax Propeller project has OSD among other things. I have been playing with the Arduino now. Been using Sparkfun $25 ArduPilot.

I would like to help here with a very reasonable priced 6DOF.
I also have the other 8 steps for the 6DOF if there is interest.
Earl

[Mikro]

Has anyone done anything with the Sparkfun 6DOF Atomic yet ?

I had previous tested with the LISY gyro but got terrible drift problems when I used my video equipment on the quad. No matter where I mounted it, for some reason the quad would drift a lot when my 900MHz wireless video system was on. Also, I read about vibration issues with it, but it looks like guys are starting to use silicon anti vibration mounts for the motors and flight boards, so thjat might not be much of an issue. Since I want to take some AP with my quad, I have stayed away from the LISY gyros.

I have developed a 6DOF for < $35 using Wii nunchuck and Wii motion plus. Works better than the Sparkfun 6DOF !

That is great! Thanks for sharing the picture. I'm sure the quaduino thread folks at RC Groups would be very interested in this, I'd suggest sharing your work there as well!

[fer]

Sorry for my English

Is it really possible to use Wii nunchuck and Wii motion plus?
I would appreciate the rest of the steps
Thank you very much :P

[Mikro]

Is it really possible to use Wii nunchuck and Wii motion plus?
I would appreciate the rest of the steps

Welcome! Yes, it should be technically possible to use these sensors. Unfortunately I've never pursued this so I wouldn't know the exact steps to follow. I'd suggest posting here as others have definitely taken that road: http://www.rcgroups.com/forums/showthread.php?t=1066119

[kast]

There seems to be some excellent results with this IMU on quadrotors.

I anyone capable of integrate the low-cost 6DOF nunchuck-WP+ into the quadcopter code ?


Kast

[Mikro]

That's great news! Any web reference we can read up on to learn from their experiences? There is a set of new ST gyros that I'm about to experiment with (I believe same vendor as the gyros in Razor). They have a yaw gyro that is in a flat package that I really like, when you buy the sensors separately with accelerometer, it comes out a little cheaper than Razor.

[lamarche_mathieu]

Hello, i'm french, and i build an aeroquad with Wii parts (Wii Motion Plus and Nunchuk).

I've just integrate the WMP+ and Nunchuk code into the firmware, i see datas into the LabView App, but i don't know the value range attempt by the software.
the accelero data are coded in 9 bits (when horizontal, accel Z = 100)
the gyro data are coded in 14 bits. when datas are received all 19ms, for 360° we get an accumulator at 5000

Please, how can i transform my datas to be compatible with the original code ?

[kast]

Hi

I've some plans to build an IMU based on a WM+ and nunchuck, however there seems to be some problems.

How much time does it take your IMU to read the values ?

Can you put some pictures ?

Thanks

Kast

[Mikro]

@Lamarche_Mathieu, the gyro roll, pitch, yaw and accel roll and pitch values are all centered at zero. This means that when the quad is motionless, take a reading. This will be your zero value. Then the code will use (current value) - (zero value) to control the quad. You can then just adjust your PID values to make sure the correct motor values are sent to the motors. (Hope that makes sense, please ask more questions if not).

I know I need to write an official Z accelerometer calibration to make it more accurate. BUT, I use atan2 to calculate angle, so as long as the X or Y axis has the same units as the Z axis, you will be OK. In the code I record the smallest Z accel and larges Z accel and the software calculates the center and assumes that is the zero value. Please keep asking more questions and I'll try to clarify it more if it's still confusing.

By the way, 19ms would be too slow a response for gyro data. With the analog sensors we use we can get data back from all 6 channels in less than a ms. 19ms is roughly 50Hz. In my initial experiments you can get a quad flying somewhat at a 50Hz rate, but stability really kicks in around 100Hz. I vaguely remember a user talking about the Nunchuck at the RC Groups Quaduino Forum (http://www.rcgroups.com/forums/showthread.php?t=1066119). You might ask your question there too.

[lamarche_mathieu]

Thank you for your response.
for the 19ms, i must re-test.
Can you explain me where accelero and gyro value are used, to understand the PID.
Thanks you.

PS : i think it's a good idea to include the WMP/Nunchuk code into the official code, because it's very easy and cheap to use Wii controler : WMP : 15$, Nunchuk : 19$ => 34$ for 6 degree.