TY - GEN
T1 - A generalized algorithm for continuous-time distributed optimization
AU - Yu, Hao
AU - Dhullipalla, Mani H.
AU - Chen, Tongwen
N1 - Publisher Copyright:
© 2021 American Automatic Control Council.
PY - 2021/5/25
Y1 - 2021/5/25
N2 - This letter proposes a new generalized continuous-time distributed optimization algorithm which includes the popular modified-Lagrangian-based (MLB) algorithm and zero-gradient-sum (ZGS) algorithm as its special cases. The convergence of the proposed algorithm to the optimal point is analyzed in a uniform framework for directed and undirected communication topologies. Moreover, it is showed that by utilizing the Hessian of local cost functions, the design of algorithmic gains is made independent of global information even if the ZGS constraints are not satisfied. Finally, numerical simulations are provided to illustrate the feasibility of the theoretical results.
AB - This letter proposes a new generalized continuous-time distributed optimization algorithm which includes the popular modified-Lagrangian-based (MLB) algorithm and zero-gradient-sum (ZGS) algorithm as its special cases. The convergence of the proposed algorithm to the optimal point is analyzed in a uniform framework for directed and undirected communication topologies. Moreover, it is showed that by utilizing the Hessian of local cost functions, the design of algorithmic gains is made independent of global information even if the ZGS constraints are not satisfied. Finally, numerical simulations are provided to illustrate the feasibility of the theoretical results.
UR - http://www.scopus.com/inward/record.url?scp=85111940034&partnerID=8YFLogxK
U2 - 10.23919/ACC50511.2021.9483292
DO - 10.23919/ACC50511.2021.9483292
M3 - Conference contribution
AN - SCOPUS:85111940034
T3 - Proceedings of the American Control Conference
SP - 820
EP - 825
BT - 2021 American Control Conference, ACC 2021
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2021 American Control Conference, ACC 2021
Y2 - 25 May 2021 through 28 May 2021
ER -