Emerging Combined Soft Computing Technologies

Fuzzy logic [161] has been successfully applied in many areas, such as control of industrial processes, control of robotic manipulators, control of servo-motors, complex decision making, diagnostic systems and others [10,93,94,98]. When using fuzzy logic, input data in form of linguistic values are represented by membership functions which are used for defining fuzzy sets of crisp values and their corresponding membership degrees related to these sets. Many parameters of systems based on fuzzy logic should be defined with help of an expert. In the same time, however, the associated parameters construction processes are performed by the method of trial and error or some heuristic algorithms [143]. Moreover, a designer that knows the characteristics of the system also needs initial rules. Some researchers have suggested a number of mechanisms to generate fuzzy rules and developed methods of their modification on the basis of experience [131,134,140,141,142]. Among them we must distinguish the self-organizing fuzzy controller that is capable of forming and modifying fuzzy rules [134,142], the clustering algorithm for fuzzy partitioning the input data space [140], and the least square algorithm for defining a succession of parameters [140,141] that can be used to construct systems based on fuzzy logic.