Thread: Visualizing 3D rotation using quaternion data

Hi all,

Just bought a new ST Micro iNEMO-M1 system, and am looking to have my code visually show me the rotation information. I've got a working C++ code which actively acquires data from the sensor, and I'm looking to extend that into some visual interface which will show me on-screen the position of the sensor.

I figure using rotation information in the form of quaternion data is the most helpful (want to avoid the gimbal lock problems associated with Roll/Pitch/Yaw).

I wanted to know if you all had any suggestions about how to get started--what would be the best method to visualize the rotation of my sensor in real-time?

Thanks!

Not sure what problem you're trying to solve here.
I tried to do something similar a few months ago (using an accelerometer).

I think your easiest path would be to use a library like glm apply the quaternion transformations on ..say a cube.
You can then render the cube anyway you like.
(AFAIK, implementing quaternion maths yourself on the gpu (..say...by glsl) isn't really recommended)

Exactly, basically just trying to get real-time feedback as to the sensor's position--so in theory a cube (or some other object--doesn't really matter) would rotate on screen matching the 3D rotation of the sensor as I rotate it in my hand.

The system outputs gyroscope, accelerometer, and magnetometer data, and the MCU uses an extended Kalman filter algorithm to convert that to Roll/Pitch/Yaw as well as quaternion data, all of which I have easy access to. Just trying to figure out the most efficient way to utilize that data to visually display the cube's 3D orientation.

The math involving quaternions definitely gets a little heavy--but it seemed like the most accurate method since it avoids that gimbal lock problem