Hey guys. I just noticed this thread and don't know if anybody is still interested, but the standard XSens MTi/x IMUs have an RS-232 interface, so they can be interfaced with the Arduino using an appropriate logic converter like the Max232 or the Sparkfun RS-232/TTL adapter board. They run at 5v, so you'll need a step-up circuit if you're interfacing with one of the 8MHz Arduinos and/or running off of a 3.7v lipoly battery. The XSens low-level com protocol is documented but the documentation can't be shared legally. But if you have the hardware, you have the documentation. Also, the math is not the same as that used on an Ardupilot or the other open source projects. The MT-series devices host a much more powerful 600MHz Blackfin DSP running their proprietary extended Kalman Filter which is individually calibrated for each unit before it is shipped, and can output very accurate Eulers, rotation matrix, or quaternion values at very high frequency. That said, for the purposes of a DIY UAV, Bill's DCM method running on the Sparkfun 9DOF is only marginally inferior. Because of the way Bill and Jordi's code weights reliability of pitch and roll data based on the magnitude of the total acceleration vector, and the magnetometer's dependence on those values for tilt compensation, it can take a little while for the yaw drift compensation to kick in when the IMU is simultaneously being subjected to large linear accelerations. Anyhow, if you were interested in using the XSens back when this thread was current, I assume that you've found your way to the 9DOF by now.
I don't have any documentation for the XSENS (one could try the website) but I know that the interface is very simple (like NMEA). I'm not sure about magnometer but I know that one has to get rid of the acceleration affect on it with some maths. The XSENS is pricey (think £10,000) and mostly the arduIMU and the other open source ones will soon match it (the maths is the same).
Well I have all the SDK and manual if you are interested I could email the coms info, I am not sure whether the arduino will process the data fast enough so it.
Comments
i am sorry but i'm not okay with what you just said:
i guess it is okay when you are talking about an airplane (moving all the time)
but it is quite different when your dealing with helicopters platform,
i guess you'll need some magnetometer to get rid of the drift on the yaw axis!
i guess that is why i asked for a documentation,
thank you anyways Giles!
i'm really interrested to get infos on the Xsens Imu,
it seems like you know much about it !
i not sure i understand the drift thing , i thaught with the 3 axis magnometer it was absolutly drift free??
do you know were i can buy one?? (maybe not a new one)
can i ask you to send me the manual and protocol ??
thank's in advance
fefenin
Paul
-
1
-
2
of 2 Next