Forum Discussion
New Input project
Hello, I thought I might take a moment or two to show something that i just started working on, that might end up being a pretty good VR controller in the end...well that is the goal so let's all...
My PM seems to stay in the Outbox, so here it goes as a post, sorry to go off-topic here, I will edit if necessary.
------------------------
Hi, thanks for agreeing to have a look at the FreePIE Magwick implementation :)
The current implementation is this one https://github.com/AndersMalmgren/FreeP ... usFuser.cs
But it creates this result for me https://www.youtube.com/watch?v=WFjsz1T ... lpage#t=97
CyberVillain referred to this as the source of that code https://github.com/xioTechnologies/Open ... onyAHRS.cs
I understand this is essentially an attempt to implement the Magwick AHRS algorithm http://www.x-io.co.uk/open-source-imu-a ... lgorithms/
I understood that for a non-magnetometer solution, the script can be even simpler, see page 29 of the Magwick paper http://www.x-io.co.uk/res/doc/madgwick_ ... report.pdf
In my opinion, maybe the best initial approach would be to have that simple algorithm as a script in FreePIE, basically taking as input the raw gyro and accelerometer data and outputting the accelerometer-corrected gyro data for pitch and roll, rather than integrating it within the build. That would enable everyone to tweak it to their liking without the need for a separate build. My problem is that my Python is nowhere near sufficient to do that.
As a side note, in the meantime, I had experimented with GlovePIE on my own super generic fusion, which I published here http://www.mtbs3d.com/phpBB/viewtopic.php?f=139&t=20626 and made a video of here http://youtu.be/jPP8XQRqlLA
Thanks for having a look at your leisure and only if you want to invest your time beyond your other awesome projects :)
------------------------
Hi, thanks for agreeing to have a look at the FreePIE Magwick implementation :)
The current implementation is this one https://github.com/AndersMalmgren/FreeP ... usFuser.cs
But it creates this result for me https://www.youtube.com/watch?v=WFjsz1T ... lpage#t=97
CyberVillain referred to this as the source of that code https://github.com/xioTechnologies/Open ... onyAHRS.cs
I understand this is essentially an attempt to implement the Magwick AHRS algorithm http://www.x-io.co.uk/open-source-imu-a ... lgorithms/
I understood that for a non-magnetometer solution, the script can be even simpler, see page 29 of the Magwick paper http://www.x-io.co.uk/res/doc/madgwick_ ... report.pdf
In my opinion, maybe the best initial approach would be to have that simple algorithm as a script in FreePIE, basically taking as input the raw gyro and accelerometer data and outputting the accelerometer-corrected gyro data for pitch and roll, rather than integrating it within the build. That would enable everyone to tweak it to their liking without the need for a separate build. My problem is that my Python is nowhere near sufficient to do that.
As a side note, in the meantime, I had experimented with GlovePIE on my own super generic fusion, which I published here http://www.mtbs3d.com/phpBB/viewtopic.php?f=139&t=20626 and made a video of here http://youtu.be/jPP8XQRqlLA
Thanks for having a look at your leisure and only if you want to invest your time beyond your other awesome projects :)