Optimal Control of Hybrid Systems via Measure Relaxations

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Buehrle, Etienne, Taş, Ömer Şahin, Stiller, Christoph
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908467119259648
author Buehrle, Etienne
Taş, Ömer Şahin
Stiller, Christoph
author_facet Buehrle, Etienne
Taş, Ömer Şahin
Stiller, Christoph
contents We propose an approach to trajectory optimization for piecewise polynomial systems based on the recently proposed graphs of convex sets framework. We instantiate the framework with a convex relaxation of optimal control based on occupation measures, resulting in a convex optimization problem resembling the discrete shortest-paths linear program that can be solved efficiently to global optimality. While this approach inherits the limitations of semidefinite programming, scalability to large numbers of discrete modes improves compared to the NP-hard mixed-integer formulation. We use this to plan trajectories under temporal logic specifications, comparing the computed cost lower bound to a nonconvex optimization approach with fixed mode sequence. In our numerical experiments, we find that this bound is typically in the vicinity of the nonconvex solution, while the runtime speedup is significant compared to the often intractable mixed-integer formulation. Our implementation is available at https://github.com/ebuehrle/hpoc.
format Preprint
id arxiv_https___arxiv_org_abs_2507_19210
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Optimal Control of Hybrid Systems via Measure Relaxations
Buehrle, Etienne
Taş, Ömer Şahin
Stiller, Christoph
Optimization and Control
Systems and Control
90C22, 93C10, 28A99
We propose an approach to trajectory optimization for piecewise polynomial systems based on the recently proposed graphs of convex sets framework. We instantiate the framework with a convex relaxation of optimal control based on occupation measures, resulting in a convex optimization problem resembling the discrete shortest-paths linear program that can be solved efficiently to global optimality. While this approach inherits the limitations of semidefinite programming, scalability to large numbers of discrete modes improves compared to the NP-hard mixed-integer formulation. We use this to plan trajectories under temporal logic specifications, comparing the computed cost lower bound to a nonconvex optimization approach with fixed mode sequence. In our numerical experiments, we find that this bound is typically in the vicinity of the nonconvex solution, while the runtime speedup is significant compared to the often intractable mixed-integer formulation. Our implementation is available at https://github.com/ebuehrle/hpoc.
title Optimal Control of Hybrid Systems via Measure Relaxations
topic Optimization and Control
Systems and Control
90C22, 93C10, 28A99
url https://arxiv.org/abs/2507.19210