Output Synchronization without Explicit Knowledge of Feedback Topologies
Jilie Zhang, Yingquan Zou, Guoxiang Gu · 2019
This paper develops a fully distributed control protocol for output synchronization of heterogeneous multi-agent systems without using the explicit knowledge of the feedback topologies. The agent systems are described by finite-dimensional state-space models. The desired output trajectories are generated by an exosystem for the controlled output of all the agent systems to synchronize in the presence of deterministic disturbances, generated by the same exosystem. However, the synchronization error can be measured at only one or a few local nodes, including at least one root node, with the rest relative output measurements. We propose a fully distributed control and estimation protocol, and show that the output synchronization can be achieved asymptotically, without additional knowledge of the feedback graphs, besides its being weakly connected. Our synchronization results are illustrated by a numerical example.