A distributed detection algorithm for collective behaviors in multiagent systems

In this paper, we propose a distributed detection algorithm for collective behaviors in multiagent systems. The multiagent systems may exhibit various collective behaviors, such as flocking, torus, swarm, under different agents interaction models and control parameters. To detect those collective behaviors, we design a local and distributed algorithm to estimate certain global characteristic signals in multiagent systems, such as network moments, which could serve as a feature indicator for group collective behaviors. The proposed algorithm relies on local information exchange among agents, and achieves estimation consensus if the communication network among agents is connected and bidirectional. The convergence of the proposed algorithm is rigorously analyzed. Simulation results are provided to illustrate the effectiveness of the proposed approach.