Average Consensus for Multi-Agent System with Measurement Noise and Binary-Valued Communication

This paper considers the average consensus problems of a class of multi-agent systems (MAS) with binary-valued communication. Each agent can only obtain its neighbor's binary-valued information under measurement noise because of limited bandwidth in communication channels. To seek consensus, we propose a two-scale multi-agent consensus algorithm with distributed strategy by combining state estimation and control design alternately. An exponential step size is chosen in the state estimation process and the estimation method can be proved to be asymptotically efficient. Additionally, by utilizing a distributed control law designed based on the estimates of the neighbors' states with a constant gain, we further prove that the proposed average consensus algorithm is convergent. Furthermore, the proposed average consensus algorithm is given and proved. Finally, some simulation results, which illustrate the effectiveness of the obtained results, are also given in the paper.