Hybrid systems: Partnerships in Soft Computing

The constituent components of soft computing have already been discussed [1]. These components are not competitive, but complementary, and much research is now being done into how this synergism can be exploited to reach soft computing's goal of achieving "tractability, robustness and low solution cost."[2]. It is recalled that each of the components has features to offer to a potential partnership. Systems in which such a partnership exists are known as 'hybrid systems'. Fuzzy logic offers to the partnership the important concept of computing with words. It deals with imprecision and information granularity and is an important tool for approximate reasoning. Neural networks have the capability for learning and adaption, while genetic algorithms make use of a systemised random search and are important for optimisation. The wide field of probabilistic reasoning is also employed for its techniques for the control of uncertainty [3]. The hybrid systems on which most research is currently being done are now examined briefly, in turn.