Probability-Guaranteed Consensus Control for Nonlinear Multi-Agent Systems Under Bit Flips

In this paper, the consensus control problem for discrete multi-agent systems is investigated in the presence of bit flips. First, a bit-rate allocation model with a finite number of bytes, together with an encoding-decoding scheme, is constructed to depict the constraints of bandwidth limitations. Then, a Bernoulli distribution is introduced to characterize the stochastic bit flips, while the impacts of bit flips on encoding-decoding processes are systematically analyzed by leveraging stochastic analysis theory. A novel control framework is proposed within a probabilistic ellipsoidal constraint, ensuring robustness against bit errors and nonlinear dynamics. By virtue of linear matrix inequalities (LMIs), sufficient conditions for the controller are derived to ensure probability-guaranteed consensus. Finally, the effectiveness of the designed controller is illustrated via numerical simulations, confirming its practical applicability in networked control systems with unreliable communication channels.