On the premise of linear system, the state space equation and transfer function can be transformed into each other, and the relationship is g (s) = c (si-a) {-1} b+d.

Even for MIMO, a set of transfer functions is intentionally obtained.

State space method and traditional transfer function method have their own advantages. Generally speaking, the state space method can better discuss the coupled MIMO system and internal dynamics, but it requires higher modeling accuracy. The transfer function method is more concise for siso system control, especially the root locus method and frequency domain design method, but it is more difficult for mimo.

There are several reasons why the paper is written in the form of state space. The purpose of this paper is to provide knowledge for professional readers. The two can be converted, whether written in state space or transfer function, which can be understood by the controller. Or the control law proposed in this paper is based on the method in state space, which will of course be written in the form of state space; Or mimo itself, the description of transfer function is not concise.

As for the realization in matlab, it is not necessarily a module in the transfer function. Simulink also has modules in the state space, and it can also be built by individual integrals.

In addition, even in top academic journals, transfer functions can often be seen, because there are many simple and easy-to-use problems in using transfer functions.