Continuous attractors of fuzzy coupled recurrent neural networks

Recent research on continuous attractors is focused on single neural networks. Our brain has a lot of functional regions and these regions mutually cooperate to each other and are coupled to the whole brain. Each region is a subnetwork of the brain. The question is if each region has continuous attractor, in which condition, the whole network can possess a continuous attractor? Based on this question, the emphasis of this paper is to investigate the continuous attractors of new networks formed by coupling different neural networks under the T-S fuzzy rule. By exploring the properties of the transfer function, conditions for the new coupled network to possess continuous attractor are successfully obtained. To verify the correctness and validity of the proposed conclusions, the results of this paper are further verified by simulation.