Hi

Do any of you have an idea how to solve the following system analytically,

A'(t) = a + b * A(t)^T - A(t) * c * A(t)^T ,where ^T means transpose

a, b and c are (2x2)-matrices containing constants and A(t) is also a (2x2)-matrix.

Initial condition is that A(0) = 0

I though about transforming the system in the following way,

Y(t) = (A(t) I) , where I is a (2x2) identity matrix

Y'(t) = d * Y - Y * e * Y ,where d = (b , a ; c ; 0) e = (c , 0) 0 = (2x2)-matrix with zeros

Now the system become homogeneous, but Y is not invertible, so it doesn't get me any further.

If the original system was homogenegous I think I would be able to solve it by transforming it to a linear ODE system and solve that one,

then transform the solution back.

I have already looked at some early litterature on the topic "matrix riccati differential equations", by Reid (1946), Levin (1959) and Coles (1965), but I didn't find it very helpful

Any suggestions on how to solve it or a referrence to some litterature describing this situation?