Why Fuzzy Logic Systems are difficult to design:
First and foremost, when designing a Fuzzy Logic System, one must carefully select Membership Functions that accurately represent the data at hand. If the membership functions do not properly reflect the data, then the entire system may be thrown off balance. Secondly, the rules which dictate the behavior of the system must be specified in a manner which is both clear and concise. Ambiguity can often lead to confusion and frustration amongst users. Lastly, the system as a whole must be designed to be sturdy against changes in input data- if not, the system may produce undesirable results.
There are many ways that AI can be used to assist in the design process of a Fuzzy Logic System. For example, AI can be used to automatically generate membership functions according to pre-defined criteria. Additionally, AI can be used to create rules for the system. Furthermore, AI can be used to monitor and assess the performance of the system on an ongoing basis. Finally, AI can be used to help design more user-friendly interfaces for Fuzzy Logic Systems- making them more accessible and intuitive for those who are less experienced.
References:
https://www.sciencedirect.com/science/article/pii/S0898122100901189
https://www.geeksforgeeks.org/design-of-fuzzy-logic-controller-using-ai-techniques/
https://www.researchgate.net/publication/220713817_A_ survey_of_current_trends_in_the_design_of_fuzzy_logic_systems
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.456.1281&rep=rep1&type=pdf
https://www.intechopen.com/online-first/fuzzy-logic-systems-design-and-applications