A Fixed-Time Consensus Control With Prescribed Performance for Multi-Agent Systems Under Full-State Constraints

This paper investigates a fixed-time consensus control problem of nonlinear multi-agent systems under full-state constraints. First, by designing corresponding constraint functions for system transformation, state-dependent asymmetric time-varying constraints are realized. The feasibility conditions of the system are eliminated, and the requirements on the constraint boundary are relaxed. Meanwhile, a prescribed performance function is designed for the constraints on synchronization deviation, which helps to improve the transient and steady-state performance of the system and ensure rapid consensus convergence on a fixed-time framework. Additionally, considering the frequent communication between the controller and actuator and to decrease the controller update frequency to save system bandwidth, an adaptive threshold event-triggered mechanism is developed. A dynamic parameter is introduced into the triggered mechanism to adjust the triggered threshold, thereby overcoming the issue that static parameters might cause excessive or insufficient event-triggered and avoiding Zeno behavior. Finally, the effectiveness of the proposed strategy is verified through simulation.
Note to Practitioners
—In complex modern engineering systems, the consensus control of multi-agent systems has become a research hotspot. In practical industrial applications, given the requirements for safety and production efficiency, it is crucial for systems to effectively constrain states and ensure performance. This study employs the prescribed performance strategy within the fixed-time framework to construct the control method. The constraints of the system’s full states are achieved by transforming a constrained system into an unconstrained one. Meanwhile, the use of event-triggered mechanisms saves communication resources. The proposed method not only ensures rapid consensus convergence under full-state constraints but also enhances the control performance of multi-agent systems, closely connected to the needs of practical applications. Future research will continue to investigate how to apply it to practical engineering applications.