How to solve an equation with two matrices?

I am trying to solve for all the unknowns in a matrix where I know the output of the matrix. This is what I mean: I have X and A and based on the two I want to find theta1 alpha d a

syms theta1 alpha d a
X = [ nan nan -3/4 1;
      nan 1/4 nan sqrt(3);
      0 nan nan 1/3;
      0 0 0 1]

A = [ cos(theta1) -sin(theta1)*cos(alpha) sin(theta1)*sin(alpha) a*cos(theta1);
      sin(theta1) cos(theta1)*cos(alpha) -cos(theta1)*sin(alpha) a*sin(theta1);
      0 sin(alpha) cos(alpha) d;
      0 0 0 1]

idx = find(~isnan(X));

solve(X(idx)==A(idx), [ theta1 alpha d a])

[Torsten]

And what is the error message you get ?

[Jimmy Neutron]

I get the output theta1 =theta1, alpha=alpha, d=d and a=a...

[Torsten]

S = solve(X(idx)==A(idx), [ theta1 alpha d a])

What is the MATLAB class of S ?

Answer 1

[Gargi Patil]

Hi,

As suggested in the comments, the result can be stored in a variable S. S will be a struct with the fields theta1, alpha, d and a as shown below:

syms theta1 alpha d a
X = [ nan nan -3/4 1;
      nan 1/4 nan sqrt(3);
      0 nan nan 1/3;
      0 0 0 1];
A = [ cos(theta1) -sin(theta1)*cos(alpha) sin(theta1)*sin(alpha) a*cos(theta1);
      sin(theta1) cos(theta1)*cos(alpha) -cos(theta1)*sin(alpha) a*sin(theta1);
      0 sin(alpha) cos(alpha) d;
      0 0 0 1];
idx = find(~isnan(X));

S = solve(X(idx)==A(idx), [theta1 alpha d a])

S = struct with fields:
    theta1: [2×1 sym]
     alpha: [2×1 sym]
         d: [2×1 sym]
         a: [2×1 sym]

You can access the possible solutions as follows:

S.a(:, 1)

ans =