Sampled-Data-Based Bipartite Leader-Follower Synchronization of Cooperation-Competition Neural Networks via Interval-Scheduled Looped-Functions

This paper addresses the bipartite leader-follower synchronization (BLFS) of cooperation-competition neural networks (CCNNs) via sampled-data (SD) control. First, the directed signed graph (SG) theory is applied to describe the cooperation and competition interactions, and thus, an appropriate mathematical model is constructed for such kind of multiple CCNNs. Based on the coordinate transformation technique, the main challenges encountered from the Laplacian matrix of the directed SG are circumvented. Then, a tractable error system can be established in the presence of SD control. Two interval-scheduled looped-functions are well-structured by relaxing the requirements of positive definiteness and continuity, respectively. In combination with discrete Lyapunov theory and some inequality techniques, some less conservative criteria are derived to guarantee the BLFS of CCNNs. Compared with the previous analysis methods, a smaller coupling strength or a larger sampling interval can be permitted on the basis of the developed results. Finally, two examples are presented to verify the effectiveness and advantages of the obtained results.