Event-Triggered Distributed Optimization Algorithm over Directed Networks: A Nonsingular Estimator Approach

This paper investigates the event-triggered distributed optimization problems (ETDOPs) over strongly connected directed networks. By assigning an additional scalar state variable to each agent and utilizing diminishing time-varying gain/step-size, a class of modified event-triggered distributed optimization algorithms (ETDOAs) is proposed, which can address the ETDOPs well and can avoid the inverse operation of some estimators in the existing literature. Compared with the existing DOAs, this paper gives a new idea to solve the DOPs under weighted-unbalanced digraphs and continuous communication of agent networks is avoided. Finally, numerical simulations are given to illustrate the effectiveness of the proposed ETDOAs.