This study has concentrated on fuzzy logic controllers from the basic aspects to an advanced self-tuning strategy. Fuzzy logic provides a very good technique for knowledge representation which makes it possible to incorporate the experience of human operators in the design of controllers;The basic concepts of fuzzy set theory, fundamental definitions of fuzzy logic, and basic structure of fuzzy logic controllers are introduced and a guideline for building the fuzzy rule-based system is developed. The rule development and adjustment strategies for fuzzy logic controllers are presented and experimentally identified. The fuzzy logic control system is analyzed on a linguistic plane with a performance trajectory. Non-linear multilevel relay property is indicated as the intrinsic feature of fuzzy logic controllers. The computer simulations and laboratory experiments indicate that the fuzzy logic controllers perform better than conventional PID controllers;Fuzzy model identification is the base to establish the initial rule set for self-tuning fuzzy logic controllers. It includes delay time determination and fuzzy parameter estimation. An artificial neural network is used as a mapping function to determine the delay time for a HVAC plant. The neural network has a feed-forward data flow mode and uses General Delta Rule as a back-propagation learning algorithm. In this study, an acceleration technique is proposed to improve the General Delta Rule. A linguistic identification method is developed to estimate fuzzy parameters for the controlled plant;A self-tuning strategy is proposed in this study as an extension of simple fuzzy logic controllers to avoid the laborious task of adjusting fuzzy logic controllers. A desired optimal performance trajectory is used as a control model. The deviation of actual performance trajectory from the desired one is used to improve the fuzzy logic controllers. The detailed performance measurement and modification procedure are developed. The self-tuning algorithm has been successfully verified in experiments.