How can I fix this error?

theta=30
sign=(cosd (theta)/abs (cosd (theta)))
u=.8
R=12
c=5
y=8
Id=(.5*Md*R^2)

Mp=75

Mb=30

alpha=0
u=.8
L=6*R

a=[1 0 1 0 0 0 0 0 0 0 0; 0 1 0 1 0 0 0 0 0 0 0; 0 0 (-R*sind (theta)) (R*cosd (theta)) 0 0 0 0 0 0 1; 0 0 0 0 1 0 1 0 0 0 0; 0 0 0 0 0 1 0 1 0 0 0; 0 0 0 0 (-R*sind (theta)) (R*cosd (theta)) 0 0 0 0 0; 0 0 -1 0 -1 0 0 0 (4*abs (sign)) 0 0; 0 0 0 -1 0 -1 0 0 (-4*u*sign) 0 0; 0 0 -c (3*R) -c  (-3*R) 0 0 (-4*(y-c)) 0 0; 0 0 0 0 0 0 0 0 (-u*sign) 1 0]

c=[0; Md*g; Id*alpha; 0; Md*g; Id*alpha; Mp*ax; Mp*y; 0; Mb*ab; Mb*g]

x=c/inv(a)
Trial>> x=c/inv(a)
Error using inv
Matrix must be square.

[John D'Errico]

Writing c/inv(a) is wrong for several reasons, sufficiently so that I have no idea what you really want to solve. To start with,

- Md is undefined

- a is not square, so you cannot form the inverse matrix

- Since / will factor the matrix inv(a) and THEN try to do something similar to taking an inverse, you are essentially trying to form the inverse of the inverse of a.

I think you need to think about what you are trying to do. At least, explain it, because your code makes no sense.

[Emily Gobreski]

Matrix [a][x]=[c]

I am trying to solve for the unknown matrix x.

Mb should be Md

Answer 1

[John D'Errico]

So x is a VECTOR, not a matrix.

You need to understand there will NOT be a unique solution. The matrix a is 10 by 11, so the problem is under-determined. There are infinitely many solutions to such a problem.

You can solve the problem A*x=c, for unknown vector x a:

x = a\c;

This will generate ONE solution, that has the property that at least one of the unknown elements in x will be zero.

Alternatively, you can use

x = pinv(a)*c;

This solution will have the property that it will have a minimum norm among all possible solutions.

And of course, you need to consider if there should be another row in the matrix a (and one more element in c). Such an under-determined problem may indicate that your problem is not formulated properly as written.