ADMISSIBLE CONSENSUS OF MULTI-AGENT SINGULAR SYSTEMS

Consensus problems are studied for both continuous-time and discrete-time multi-agent singular systems with time-invariant and directed communication topologies. Under restricted system equivalence of singular agents, sufficient and necessary conditions are obtained for admissible consensus ability with static protocols, which are based on both the relative information of the dynamic states and the absolute information of the static states. For a network of continuous-time singular systems, the existence of admissible consensualization can be cast into strong stabilizability of the agent dynamics. Once discrete-time multi-agent singular systems satisfy the condition of reaching nontrivial final consensus states, strong stabilizability is a sufficient condition to achieve admissible consensualization. Two algorithms are proposed to construct two protocols, which are based on a linear matrix inequality and a modified Riccati equation, respectively. Finally, the algorithms are illustrated by two simulation examples.