Design and analysis of quantizer for multi-agent systems with a limited rate of communication data

This paper concerns the problem of consensus for multi-agent systems with a limited rate of communication data. The effect of the quantization error on consensus is analyzed for the case that the nodes’ dynamics is described by the first order integrator with a quantized message input. In order to describe how exactly the consensus is achieved, the notion of consensus level is introduced. Based on these, a suboptimal method of resource allocation to reduce such effect of quantization error is proposed, and a dynamic quantizer is designed to meet the requirement of the consensus level, under the constraints that the total quantization rate of overall system is limited. At last, a numerical example is included for demonstration, and the outcome of the simulations suggests the validity of the presented theoretic results.

► We design a quantizer based on the initial relative states of agents.
► The quantizer reduces the final disagreement among agents effectively.
► Adjusting the quantization step size will also reduce the final disagreement.