Appointed-weight Privacy-preserving Consensus of Multi-agent Systems

Though privacy-preserving consensus of multi-agent systems has been extensively studied, one issue remains unsolved. That is, the final consensus value may not be exactly the average of all the agents' initial values. In this paper, we investigate the privacy-preserving consensus problem of multiagent systems, where the final consensus value is the weighted average of all the agents' initial values with predefined weights. A noise generation mechanism is designed, and a novel privacy-preserving consensus algorithm is proposed. Theoretical analysis shows that the proposed algorithm can realize appointed-weight consensus while achieving privacy preserving.