Distributed ADMM Optimization Algorithm with Communication Delay

This paper investigates the ADMM algorithm to solve the problem of distributed optimization of multi-agent systems (MASs) with communication delay. The communication delay leads to state estimation errors coupled with the state variables, which makes it difficult for the convergence analysis of the ADMM algorithm. We’re introducing a new distributed algorithm which can achieve linear convergence when local cost functions are strongly convex and their gradients are Lipschitz. Finally, an illustrative example demonstrates the derived results.