To design a controller for nonlinear and uncertain systems we have two choice: the first one is fuzzy sliding mode controller and the second one is sliding mode fuzzy controller. I'd like to know that which one is better and why?
As far my knowledge, there is nothing strongly specific called 'fuzzy Sliding mode controller' or "Sliding mode fuzzy controller". It is dependent on the designer, how he used the theme. [need to stydy his method].
To me, if any 'blend' between fuzzy [as a tool] and SMC should termed as 'fuzzy Sliding mode controller'. SMC is a robust nonlinear controller for plenty of applications. In most cases fuzzy tool is added with SMC to achieve specific task like decresing parameter uncertainty, increasing stability and performance.
If the fuzzy act as the controller or something like that, ___ -fuzzy controller may exist. but I have seen that fuzzy is used as a tool in all "Sliding mode fuzzy controller" themes.
Fuzzy sliding mode controller (FSMC) is a nonlinear controller based on sliding mode method when fuzzy logic methodology applied to sliding mode controller to reduce the high frequency oscillation (chattering) and compensate the dynamic model of uncertainty based on nonlinear dynamic model.
Sliding mode fuzzy controller (SMFC) is an artificial intelligence controller based on fuzzy logic methodology when, sliding mode controller is applied to fuzzy logic controller to reduce the fuzzy rules and refine the stability of close loop system in fuzzy logic controller.
However the SMFC has a good condition but the main drawback compared to FSMC is calculation the value of sliding surface slope coefficient pri-defined very carefully and FSMC is more suitable for implementation.
Before I wrote my paper "Slinding mode fuzzy control" I was wondering why a diagonal Mamdani FC with rules like IF error=PB and change of error=NS THEN u=NB work so well and robustly with regard to uncertainties even for a nonlinear system whose parameters are not completely known in advance. The reason is that this kind of diagonal FC works as an SMC with a nonlinear boundary layer. However the main point is to choose the input scaling factors for error and change of error and the output gain accordingly so that stability, performance and robustness are guaranteed. In the following this SMFC was enhanced by the equivalent control which serves as a kind of compensation term for well known parts of the system.
In my opinion, now a controller is called "FSMC" for all fuzzy controllers with a sliding mode term included which is also the case for the SMFC. So I can't see a significant difference between the two.