Model-Free Adaptive Control for Nonlinear Multi-Agent Systems Under Constrained Bit Rate: An Encoding-Decoding Mechanism

Abstract

In this article, the model-free-based consensus tracking control problem is investigated for nonlinear discrete-time multi-agent systems (MASs) subject to constrained bit rate. A dynamic linearization technology is employed, by which the original unknown nonlinear system is equivalently transformed into a dynamic linearized model. An encoding-decoding mechanism (EDM) is applied to encode the measurement outputs into the binary codewords with fewer occupations of the network bandwidth. On this basis, a distributed model-free adaptive control (MFAC) scheme is developed, while sufficient conditions are presented to ensure that the close-loop MAS achieves the expected consensus performance. The proposed scheme is completely data-driven without relying on any information from the system model or structure. Meanwhile, the inherent relationship between bit rate constraints and decoding accuracy is revealed. Finally, simulation results are presented to demonstrate the validity of the provided approach.