Evolutionary Design of Very Compact Analog Circuit Implementations of Fuzzy Systems

Abstract

Evolutionary algorithms (EA) offer good promise for automated design of analog circuits as well as for adaptation and automatic reconfiguration of programmable devices. In particular, EA facilitate the design of analog circuits for very specific requirements, such as those related to the implementation of fuzzy operators, or even of complete fuzzy systems. The paper starts with a brief overview of the evolutionary design process and of a family of analog programmable devices that support on-chip evolution, illustrates evolutionary design with examples of evolved analog circuits implementing fuzzy parametric t-norms, and focuses on an evolved circuit with only 7 transistors, which approximates the control surface of a 2-input fuzzy controller, obtaining thus a mapping of a complete fuzzy system in a circuit with only a few transistors. The paper presents evidence that EA can provide very compact solutions for implementation of fuzzy systems, and that programmable analog devices are an efficient and rapid solution for rapid deployment of fuzzy systems.