Predefined-time consensus control for multiagent systems with input and output quantization

For the predefined-time control problem of unknown nonlinear multi-agent systems (NMASs), a novel adaptive neural consensus control strategy is proposed. Unlike existing predefined-time control approaches, this strategy enables communication of input and output signals through a directed network, and incorporates quantization prior to communication. Firstly, a neural network is employed to approximate unknown functions. Based on the quantized information, a neural-network-based distributed state observer is designed to estimate the unmeasurable states. Secondly, this paper combines backstepping technique with command filtering technology, thereby avoiding the non-existence issue of partial derivatives of virtual control signals caused by output quantization. Moreover, the intermediate auxiliary control signal is constructed using a class of smooth functions. By substituting the quantized output for the continuous output, the actual controller is obtained. To analyze the predefined-time stability of the system, it is necessary to compensate for quantization errors. Based on this, Lemma 12 is proposed. Finally, in a predefined time frame, it is proven that the outputs of the followers converge to a neighborhood of the leader's output, while ensuring all signals within the closed-loop system remain bounded.