We are running a survey to help us improve the experience for all of our members. If you see the survey appear, please take the time to tell us about your experience if you can.
Newbie to the embedded systems here ;) We have a Cortex M4 with DSP.
We've got a linear system of equations to solve. Say, we have 8 variables to solve, so vectorize it into AX=B.
So it's required that A is (n,8) sized, X=(8,1), and B=(n,1)(with n>=8).
If A is a square matrix with a rank of 8, then I could've easily gotten X by inv(A).B, where arm_mat_inverse_f32.c and arm_mat_mult_f32.c could be directly invoked.
But, if we got an A matrix with n>8 and ranked 8, how to solve the X? Is there a way to automatically do something like elementary row transformations?
Appreciate everything. Please bear with my noobness ;)