Difference between revisions of "Talk:Roadmap"

From Chordata documentation
Jump to: navigation, search
(Improve calibration process: new section)
(add 3 kalman)
Line 20: Line 20:
*[https://www.mdpi.com/1424-8220/17/9/2146/pdf TGIC-KALMAN w/ magnetic compensation] paper by Kaiqiang Feng et al (2017).
*[https://www.mdpi.com/1424-8220/11/10/9182/pdf quaternion-based extend Kalman filter for human body tracking] paper by Angelo Maria Sabatini (2011)
*Implmentation in C by Alberto Petrucci https://os.mbed.com/users/capriele/code/IMU-9DOF-KALMAN/
'''Use higher sensor refresh rates'''
'''Use higher sensor refresh rates'''

Revision as of 21:34, 5 May 2019

Please use this page to suggest new features to be implemented Anyone can add content to a wiki like this, see How_to_contribute for more information

3D preview on the browser

  • This is a feature that I would really like to add: Having a live avatar preview directly on the Remote console environment. it can be implemented with Three.js / Daylankifky (talk) 09:20, 9 April 2019 (UTC)
In order to implement it the calibration process should happen inside the SBC, a way to do it is to use the scripts from the Blender add-on on a headless Blender process / Daylankifky (talk) 09:20, 9 April 2019 (UTC)

Next hardware revision

  • Change led polarity on K-Ceptor Daylankifky (talk) 11:06, 18 April 2019 (UTC)

Improve Sensor accuracy

Implement different sensor fusion algorithms

About magnetic disturbances Dealing with Magnetic Disturbances in HumanMotion Capture: A Survey of Techniques, by Gabriele Ligorio * and Angelo Maria Sabatini

KALMAN implementations:

Use higher sensor refresh rates https://forum.chordata.cc/d/14-questions-on-fidelity/2

Daylankifky (talk) 10:18, 20 April 2019 (UTC)

Improve calibration process

Implement a different ellipsoid fitting algorithm

Provide calibration helpers on the remote-console

  • For visualizing point cloud sphere, read and corrected
  • For testing the calibration (like the testcube.blend)