deriving a position matrix

I have a position matrix of my joints like this

Joint1 [p1 p2 p3 p4 p5 p6]

Joint2 [p1 p2 p3 p4 p5 p6]

Joint3 [p1 p2 p3 p4 p5 p6]

I want to get the velocity matrix by deriving this matrix, how is this possible?

Answer 1

[Ishu]

Hi Tim,

In order to derive the velocity matrix from the position matrix you need to differentiate position matrix with respect to time.

I am assuming that the sample rate is constant, hence the velocity can be approximated by the difference between the successive position joints divided by the time interval between those joints. To calculate the difference you can use "diff" function of MATLAB.

% Replace your positionMatrxi with your actual values
positionMatrix = [
    1, 2, 3, 4, 5, 6;  % Joint1
    2, 3, 4, 5, 6, 7;  % Joint2
    3, 4, 5, 6, 7, 8   % Joint3
];

deltaT = time_difference;  % Time interval between samples
velocityMatrix = diff(positionMatrix, 1, 2) / deltaT;

You can refer this documentation for "diff" function:

https://www.mathworks.com/help/matlab/ref/diff.html#btwmxq8-3

Hope it helps!

[Tim De Witte]

Yes I did that but I would need a matrix of the same size :/

[Voss]

Look into using gradient instead of diff.