Multi-agent consensus based on the group combined information

This paper considers the multi-agent consensus problem for first-order discrete-time dynamic systems. Instead of utilizing the relative state information as a feedback, the state of every individual agent is updated based on the relative information between two groups, where the agents are randomly partitioned into two groups each time. In this paper, we consider the group information as the convex combination of all the states of the agents in each group. The convex combination with random coefficients and one case with equal probability for group partition are analyzed. In addition, the application of this new scheme to the distributed beamforming is given. Numerical simulations are also provided to demonstrate the validity of our theoretical results.