Consensus of multi-agent systems under binary-valued measurements: An event-triggered coordination approach

Abstract

This paper investigates the consensus problem of multi-agent systems. Owing to the effect of bandwidth restriction, the communication information between two adjacent agents is based on the binary-valued observation of the imprecise sender’s output, which is only 1-bit. Each agent employs the recursive projection algorithm to estimate the states of all neighbors. Then, an event-triggered coordination technique is proposed to address the consensus problem. The first characteristic of this technique is that the information data transmission between two adjacent agents is managed by an event-triggered scheduler. The second characteristic of this technique is that each agent adopts the event-triggered control protocol to achieve the ultimate target. By this technique, the estimation error and the consensus error can converge to zero in the mean square sense and in the almost sure sense with explicit convergence speeds. Moreover, the communication rate of the communication scheme between two adjacent agents does not exceed fifty percent. And the communication rate of the control scheme can almost surely converge to zero with an explicit rate. Compared with the existing results, these two communication rates can be extremely reduced when achieving the same convergence value. As a result, the communication resources can be saved in terms of quantization, communication and control. An example is presented to illustrate the advantage of the proposed technique.