On Convex Data-Driven Inverse Optimal Control for Nonlinear, Non-stationary and Stochastic Systems

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Garrabe, Emiland, Jesawada, Hozefa, Del Vecchio, Carmen, Russo, Giovanni
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917706128687104
author Garrabe, Emiland
Jesawada, Hozefa
Del Vecchio, Carmen
Russo, Giovanni
author_facet Garrabe, Emiland
Jesawada, Hozefa
Del Vecchio, Carmen
Russo, Giovanni
contents This paper is concerned with a finite-horizon inverse control problem, which has the goal of reconstructing, from observations, the possibly non-convex and non-stationary cost driving the actions of an agent. In this context, we present a result enabling cost reconstruction by solving an optimization problem that is convex even when the agent cost is not and when the underlying dynamics is nonlinear, non-stationary and stochastic. To obtain this result, we also study a finite-horizon forward control problem that has randomized policies as decision variables. We turn our findings into algorithmic procedures and show the effectiveness of our approach via in-silico and hardware validations. All experiments confirm the effectiveness of our approach.
format Preprint
id arxiv_https___arxiv_org_abs_2306_13928
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle On Convex Data-Driven Inverse Optimal Control for Nonlinear, Non-stationary and Stochastic Systems
Garrabe, Emiland
Jesawada, Hozefa
Del Vecchio, Carmen
Russo, Giovanni
Optimization and Control
Information Theory
Machine Learning
Robotics
Dynamical Systems
This paper is concerned with a finite-horizon inverse control problem, which has the goal of reconstructing, from observations, the possibly non-convex and non-stationary cost driving the actions of an agent. In this context, we present a result enabling cost reconstruction by solving an optimization problem that is convex even when the agent cost is not and when the underlying dynamics is nonlinear, non-stationary and stochastic. To obtain this result, we also study a finite-horizon forward control problem that has randomized policies as decision variables. We turn our findings into algorithmic procedures and show the effectiveness of our approach via in-silico and hardware validations. All experiments confirm the effectiveness of our approach.
title On Convex Data-Driven Inverse Optimal Control for Nonlinear, Non-stationary and Stochastic Systems
topic Optimization and Control
Information Theory
Machine Learning
Robotics
Dynamical Systems
url https://arxiv.org/abs/2306.13928