Preconditioning Matrix Synthesis for a Projected Gradient Method for Solving Constrained Linear-Quadratic Optimal Control Problems

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

2 Downloads (Pure)

Samenvatting

This paper presents a method for synthesizing preconditioning matrices for a heavy-ball accelerated projected primal-dual method. The main focus lies on linear quadratic optimal control problems, as they have a specific structure that can be exploited for fast computational times. This gradient method is rewritten into a Lur’e-type system, such that convergence of the algorithm can be enforced through finding an appropriate Lyapunov function for the Lur’e system. It has been shown that for a small problem, it is possible to synthesize preconditioning matrices and that the method is 10^4 times faster than solving the projection using a dedicated solver.
Originele taal-2Engels
Titel2023 62nd IEEE Conference on Decision and Control, CDC 2023
UitgeverijInstitute of Electrical and Electronics Engineers
Pagina's7253-7258
Aantal pagina's6
ISBN van elektronische versie979-8-3503-0124-3
DOI's
StatusGepubliceerd - 19 jan. 2024
Evenement2023 62nd IEEE Conference on Decision and Control (CDC) - Singapore, Singapore, Singapore, Singapore
Duur: 13 dec. 202315 dec. 2023
Congresnummer: 62

Congres

Congres2023 62nd IEEE Conference on Decision and Control (CDC)
Verkorte titelCDC 2023
Land/RegioSingapore
StadSingapore
Periode13/12/2315/12/23

Vingerafdruk

Duik in de onderzoeksthema's van 'Preconditioning Matrix Synthesis for a Projected Gradient Method for Solving Constrained Linear-Quadratic Optimal Control Problems'. Samen vormen ze een unieke vingerafdruk.

Citeer dit