MATLAB: Is it possable to plot the seperate pods of the Myo armband using the MEX Wrapper

MATLABmexmyo armbandplotting

I have been using the MEX wrapper to integrate the Myo Armband with MATLAB. Using MEX I have been able to stream the EMG, accelerometer, gyroscopic, and pose data. I can then plot all of this data nice and pretty with the plot function.

plot(emg_data.time_log, emg_data.emg_log, '-');

However, I really need to plot each EMG sensor pods separately. Is there a nice way to do this?

Best Answer

Okay I feel dumb. Found the answer myself by just inspecting the collected data. To plot the individual EMG pods you need to just reference the EMG log column. For example:

plot(emg_data.time_log, emg_data.emg_log(:,1), '-');

Where column 1 would be pod zero, 2 would be pod 1 and so on. Hope this someone else.

Related Solutions

MATLAB: Mex/Fortran passing data

Download this package from the FEX:

http://www.mathworks.com/matlabcentral/fileexchange/25934-fortran-95-interface-to-matlab-api-with-extras

Follow the instructions to compile the modules. Then you can do stuff like this:

m-code:

myFortranMexRoutine(AeroConfig_Hub_position,...
                    AeroConfig_Hub_orientation,...
                    AeroConfig_Hub_translationvel,...
                    AeroConfig_Hub_rotationvel);

Fortran code:

#include "fintrf.h"
subroutine mexFunction(nlhs,plhs,nrhs,prhs)
use MatlabAPImx
implicit none
integer nlhs, nrhs
mwPointer plhs(*), prhs(*)
__yourtype__ Aeroconfig
AeroConfig%Hub%position = fpGetPr1(prhs(1))
AeroConfig%Hub%orientation = fpGetPr(prhs(2))
AeroConfig%Hub%translationvel = fpGetPr1(prhs(3))
AeroConfig%Hub%rotationvel = fpGetPr1(prhs(4))

Caution: The above is just an outline with no error checking. In production code you would need to check nrhs and nlhs to see if there are the correct number of inputs & outputs. And you would have to retrieve the data pointers via fpGetPr1 and fpGetPr into separate variables and examine if they are associated prior to using them. Etc.

MATLAB: Calculate the overlap to validate model

I created two signals to experiment with this. The code is straightforward:

t = linspace(0, 100, 1000);
EMG = sin(2*pi*t/20) .* cos(2*pi*t/50) + 1;
AB = sin(2*pi*t/25 + 0.2) .* cos(2*pi*t/40) + 1;
EMGT = 1*(EMG > 0.25) + 0*(EMG <= 0.25);
ABT = 1*(AB > 0.25) + 0*(AB <= 0.25);
Active = EMGT .* ABT;
ActivePct = 100 * sum(Active)/numel(Active);
figure(1)
subplot(3,1,1)
plot(t, EMG, '-b')
hold on
plot(t, AB, '-r')
hold off
legend('EMG', 'Model')
grid
subplot(3,1,2)
plot(t, EMGT, '-c')
hold on
plot(t, ABT, '-m')
hold off
axis([xlim    -0.1  1.1])
legend('EMG', 'Model')
grid
subplot(3,1,3)
plot(t, Active, '-g', 'LineWidth',2)
axis([xlim    -0.1  1.5])
text(mean(t), 1.25, sprintf('Percent Both Active = %.2f', ActivePct), 'HorizontalAlignment','center', 'VerticalAlignment','middle')
grid

Experiment to get the result you want.

Best Answer

Related Solutions

MATLAB: Mex/Fortran passing data

MATLAB: Calculate the overlap to validate model

Related Question