Cooperative control and communication optimization for networked multi-agent systems with nonholonomic constraints

This paper presents a cooperative formation control on networked multi-agent systems, in which the mobile agents have nonholonomic constraints. To simplify the control design, the dynamics of agents are first transformed into chained form. Based on it, a formation cooperative control scheme is derived from the conventional consensus control law which has been applied to address the nonholonomic constraints. A desired distance among agents is introduced into the consensus control law to let the agent maintain certain range among them and keep formation. The distance is determined by communication quality which is described by Shannon capacity outage probability. Then a model-free extremum seeking algorithm is used to find the best distance among agents. The proposed formation scheme combing extremum seeking can optimize the communication quality of the formation for network agents with nonholonomic constraints. The simulation is provided to validate its feasibility and effectiveness.