1: \begin{abstract}

2:     This paper proposes a novel distributed event-triggered

3:     algorithmic solution to the multi-agent average consensus problem

4:     for networks whose communication topology is described by

5:     weight-balanced, strongly connected digraphs.  The proposed

6:     event-triggered communication and control strategy does not rely

7:     on individual agents having continuous or periodic access to

8:     information about the state of their neighbors. In addition, it

9:     does not require the agents to have a priori knowledge of any

10:     global parameter to execute the algorithm.  We show that, under

11:     the proposed law, events cannot be triggered an infinite number of

12:     times in any finite period (i.e., no Zeno behavior), and that the

13:     resulting network executions provably converge to the average of

14:     the initial agents' states exponentially fast. We also provide

15:     weaker conditions on connectivity under which convergence is

16:     guaranteed when the communication topology is switching.  Finally,

17:     we also propose and analyze a periodic implementation of our

18:     algorithm where the relevant triggering functions do not need to

19:     be evaluated continuously.  Simulations illustrate our results and

20:     provide comparisons with other existing algorithms.

21:   \end{abstract}

22: