Distributed Asynchronous Discrete-Time Feedback Optimization

In this article, we present an algorithm that drives the outputs of a network of agents to jointly track the solution of a time-varying, strongly convex optimization problem. This algorithm is robust to asynchrony in the agents' operations, namely, first, computations of control inputs, second,...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on automatic control Vol. 70; no. 6; pp. 3968 - 3983
Main Authors: Behrendt, Gabriel, Longmire, Matthew, Bell, Zachary I., Hale, Matthew
Format: Journal Article
Language:English
Published: New York IEEE 01.06.2025
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Algorithms
Asymptotic properties
Asynchronous optimization algorithms
Computational modeling
Convexity
Current measurement
Feedback
Heuristic algorithms
Load flow
Machine learning algorithms
multiagent systems
Optimization
Perturbation methods
Robot sensing systems
time-varying optimization
Time-varying systems
Vectors
ISSN:0018-9286, 1558-2523
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this article, we present an algorithm that drives the outputs of a network of agents to jointly track the solution of a time-varying, strongly convex optimization problem. This algorithm is robust to asynchrony in the agents' operations, namely, first, computations of control inputs, second, linear measurements of network outputs, and third, communications of agents' inputs and outputs. We first show that our distributed asynchronous algorithm converges to the solution of a time-invariant feedback optimization problem in linear time. Next, we show that our algorithm tracks the solution of a time-varying feedback optimization problem within a bounded error dependent upon the movement of the minimizers and degree of asynchrony, which we make precise. These convergence results are extended to quantify agents' asymptotic behavior as the length of their time horizon approaches infinity. Then, to ensure satisfactory network performance we specify the timing of agents' operations relative to changes in the objective function that ensure a desired error bound. Numerical experiments verify these developments and show the utility of feedback optimization under asynchrony.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0018-9286
1558-2523
DOI:10.1109/TAC.2024.3520677