Distributed Asynchronous Time-Varying Quadratic Programming with Asynchronous Objective Sampling

TL;DR

This paper introduces a multi-agent optimization framework that handles asynchronous sampling, communication, and computation in time-varying quadratic programs, addressing an open problem and demonstrating bounded error in solution tracking.

Contribution

It presents the first framework allowing asynchronous sampling in multi-agent time-varying quadratic programming, with theoretical error bounds and validation through simulations.

Findings

Agents maintain bounded error in solution tracking.

The framework effectively handles asynchronous sampling, communication, and computation.

Simulations confirm theoretical results.

Abstract

Existing works on multi-agent time-varying optimization allow agents to asynchronously communicate and/or compute, but do not allow asynchronous sampling of objectives. Sampling can be difficult to synchronize, and we therefore present a multi-agent optimization framework that allows asynchrony in sampling, communications, and computations for time-varying quadratic programs. We show that agents have bounded error when tracking the solution to the asynchronously sampled problem, which solves an open problem for quadratic programs. Simulations validate these results.