Dear Mr. Yu,

Fuzzy logic is already playing an important role. IN some research studies Fuzzy logic is used in object based image segmentation to detect if the image has normal objects or defect objects before sending the image to an Artificial Neural Network segmentation method. I am expecting that in the future this role (fuzzy logic) will be more obvious in many machine learning tasks which involves AI.

Dear Dr. Awad, Thanks for you comments, I agree with you, especially in decision-making domain of autonomous driving. The deep learning based decision-making solution actions as a black box and it is unacceptable when an accident happens but not able to give the exact reason.

It's a long story my dear ( thirty years !) . look at the works given below :

Neural networks and fuzzy systems_A Dynamical Systems Approach to Machine IntelligenceVolume: Author(s):Kosko B. Series: Periodical: Publisher:Prentice-Hall International, Inc. City: Year:1991

C++ Neural Networks and Fuzzy LogicVolume: Author(s):Valluru B. Rao, Hayagriva Rao Series: Periodical: Publisher:M & T Books City: Year:1995

Fusion of Neural Networks, Fuzzy Sets, and Genetic Algorithms: Industrial ApplicationsVolume: Author(s):Lakhmi C. Jain, N.M. Martin Series:International Series on Computational Intelligence Periodical: Publisher:CRC Press City: Year:1998

Fuzzy neural network theory and applicationVolume: Author(s):Puyin Liu,

Hongxing Li Series in machine perception and artificial intelligence 59 Periodical: Publisher:World Scientific City:River Edge, NJ Year:2004

Foundations of neural networks, fuzzy systems, and knowledge engineeringVolume: Author(s):Nikola K. Kasabov Series:American Men and Women of Science Periodical: Publisher:The MIT Press City: Year:1996