LQ and more: Matrix computations about automationshield HOT 2 CLOSED

@AnnaVargova

The EigenArduino is interesting, since the creators maintain the stlport as well. What is even more interesting, they have a Kalman filter for state estimation (also necessary for us in the future) using the EigenArduino! My first and foremost feeling is that this is a good choice...

Also, here is a full documentation of Eigen: https://eigen.tuxfamily.org/dox/GettingStarted.html

I see MatrixMath has a more "classical" and less user friendly way of handling the task. However, it is the question of which one is faster! https://playground.arduino.cc/Code/MatrixMath

According to my quick read, EigenArduino is more advanced. But MatrixMath also provides inversion. (see https://www.quora.com/Does-Arduino-has-library-to-solve-matrices/answer/Prasant-Bhatt-1). I would be interested in their memory efficiency and speed though! Here they actually mention that it is very bloated (memory wise)...

from automationshield.

Interesting forum thread on the efficiency of floating point operations on the AVR: https://forum.arduino.cc/index.php?topic=92684.0

from automationshield.