Finite-Time Convergent Algorithms for Time-Varying Distributed Optimization
From MaRDI portal
Abstract: This paper focuses on finite-time (FT) convergent distributed algorithms for solving time-varying distributed optimization (TVDO). The objective is to minimize the sum of local time-varying cost functions subject to the possible time-varying constraints by the coordination of multiple agents in finite time. We first provide a unified approach for designing finite/fixed-time convergent algorithms to solve centralized time-varying optimization, where an auxiliary dynamics is introduced to achieve prescribed performance. Then, two classes of TVDO are investigated included unconstrained distributed consensus optimization and distributed optimal resource allocation problems (DORAP) with both time-varying cost functions and coupled equation constraints. For the previous one, based on nonsmooth analysis, a continuous-time distributed discontinuous dynamics with FT convergence is proposed based on an extended zero-gradient-sum method with a local auxiliary subsystem. Different from the existing methods, the proposed algorithm does not require the initial state of each agent to be the optimizer of the local cost function. Moreover, the provided algorithm has a simpler structure without estimating the global information and can be used for TVDO with nonidentical Hessians. Then, an FT convergent distributed dynamics is further obtained for time-varying DORAP by dual transformation. Particularly, the inverse of Hessians is not required from a dual perspective, which reduces the computation complexity significantly. Finally, two numerical examples are conducted to verify the proposed algorithms.
This page was built for publication: Finite-Time Convergent Algorithms for Time-Varying Distributed Optimization
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6413307)
