Quantized Observer-Based Coordination of Linear Multi-Agent Systems

Abstract: For the coordination of multi-agent systems with bandwidth limited digital channels, if there are unmeasurable states in agents’ dynamics, then to get the estimates of other agents’ states by properly designed encoding-decoding scheme is the objective of the cooperative communication and is also the foundation of cooperative control design. In this paper, the concept of cooperatability of digital multi-agent networks is proposed with respect to quantized-observer based communication protocols and certainty equivalence principle based control protocols to characterize the cooperative communication and control in an integrative framework. The cooperatability for linear discrete-time multi-agent systems is investigated with unmeasurable states and limited data rate. Some necessary conditions and sufficient conditions are given for the existence of admissible communication and control protocols to ensure both the cooperative state observation and cooperative stabilization. It is shown that these conditions are quantitatively related to the stabilizability and detectability of agents’ dynamics and the topology of the communication network.