Adaptive fixed-time bipartite tracking consensus control for unknown nonlinear multi-agent systems: An information classification mechanism

This paper is concerned with the problem of bipartite tracking consensus for high-order unknown nonlinear multi-agent systems with actuator faults. Unlike the traditional condition that the directed signed graph is structurally balanced, a directed signed graph containing a spanning tree is considered. Besides, the consensus errors are required to satisfy both the prescribed performance and fast convergence (fixed-time). By proposing an information classification mechanism, each agent selectively uses neighbor information such that agents in the system are divided into two styles, which transform the bipartite tracking consensus problem into a general tracking consensus problem. By using neural networks and adaptive technologies to approximate unknown functions, the adaptive fault-tolerant fixed-time consensus controllers are developed. All signals in the system are bounded within a fixed time. Moreover, the bipartite consensus errors satisfy the prescribed performance by selecting appropriately predefined performance functions. Stability analysis and simulation results further verify the effectiveness of the proposed method.