Allow static gain StateSpace objects to be straightforwardly created. #107
Conversation
…ects. Allows StateSpace([],[],[],D), which failed previously. Static gains have sizes enforced as follows: A 0-by-0, B 0-by-ninputs, C noutputs-by-0. Tests added for instantiation, and sum, product, feedback, and appending, of 1x1, 2x3, and 3x2 static gains StateSpace objects.
On Python 2.7, the special case "all states useless" in _remove_useless_states resulted in A=[[0]] (and similarly for B and C). The special case is no longer needed, since empty A, B, C matrices can be handled. numpy.delete does the right thing w.r.t. matrix sizes (e.g., deleting all columns of a nxm matrix gives an nx0 matrix). Added test for this.
| raise ValueError("D must have the same column size as B.") | ||
| if self.outputs != D.shape[0]: | ||
| raise ValueError("D must have the same row size as C.") | ||
| raise ValueError("A and C C must have the same number of columns.") |
There was a problem hiding this comment.
I think it's one C too many here. 😸
|
D'oh! Thanks, will fix. |
|
I like this syntax for creating static gain objects. This PR looks good to me. |
|
One case not covered by this PR is the "empty" system, Wanting 0-input, 0-output systems may seem like fussiness, but it turns out to be necessary (or useful, at least), in a prototype of [1] http://octave.sourceforge.net/control/function/augw.html |
Conflicts: control/tests/statesp_test.py
|
Pull request #110 subsumes this one. |
Lets the (fairly?) obvious StateSpace([],[],[],D) work. One could previously create these by, e.g., converting a static gain TransferFunction to a StateSpace object.
In this case A will be 0-by-0, B 0-by-inputs, and C outputs-by-0. My intuition is that this is a generally correct approach (consider the simplification of _remove_useless_states, for example), but it may require some thought during review.