Predefined-Time Adaptive Consensus Control for Nonlinear Multi-Agent Systems with Input Quantization and Actuator Faults

Predefined-time control has experienced substantial advancements in recent years. Nevertheless, the current technology has not achieved widespread adoption in nonlinear multi-agent systems (NMASs), and significant issues pertaining to the input quantization and actuator failures remain unaddressed. This paper investigates the problem of predefined-time consensus control for NMASs with input quantization and actuator faults. Notably, the study takes into account scenarios where each actuator may experience an infinite number of faults. In conjunction with practical predefined-time stability theory, an innovative predefined-time adaptive consensus control method has been developed within the framework of the backstepping method, incorporating the approximation technique of the multi-dimensional Taylor network (MTN). Additionally, by utilizing the characteristics of quantized nonlinear sectors and the structural model of actuator faults, novel adaptive estimation techniques are devised to handle the effects caused by actuator faults and quantized inputs. To further alleviate computational burdens and tackle the issue of computational explosion, a finite-time differentiator is employed to estimate the derivative of the virtual control. The proposed control scheme achieves the desired performance of predefined-time convergence. Rigorous theoretical analyses indicate that the proposed control scheme can drive consensus errors to converge within a small range within a predefined-time, and users have the flexibility to choose the settling time. Moreover, all signals in the closed-loop system remain bounded. Finally, simulation results are provided to validate the effectiveness of the proposed approach.

Nonlinear Dynamics