Sampling-based event-triggered consensus for multi-agent systems

This work considers the event-based consensus algorithms for multi-agent systems with periodic sampling schemes. An event-triggered consensus algorithm using sampled states are proposed. The measurement error in the event design is defined based on the state of local neighborhood center in order to reduce the number of events. Based on this algorithm, a self-triggered consensus protocol is proposed to further reduce the amount of communication and state sampling, as well as the number of events and thus the controller updates. It is noted that by introducing the periodic sampling scheme, Zeno behaviors can be naturally avoided in both of the algorithms. The effectiveness of the proposed algorithms is illustrated by simulation examples.