Salvato in:
Dettagli Bibliografici
Autori principali: Aoyama, Yuichiro, Theodorou, Evangelos A.
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2403.18130
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916396927025152
author Aoyama, Yuichiro
Theodorou, Evangelos A.
author_facet Aoyama, Yuichiro
Theodorou, Evangelos A.
contents We present a sampling-based trajectory optimization method derived from the maximum entropy formulation of Differential Dynamic Programming with Tsallis entropy. This method is a generalization of the legacy work with Shannon entropy, which leads to a Gaussian optimal control policy for exploration during optimization. With the Tsallis entropy, the policy takes the form of $q$-Gaussian, which further encourages exploration with its heavy-tailed shape. Moreover, the sampling variance is scaled according to the value function of the trajectory. This scaling mechanism is the unique property of the algorithm with Tsallis entropy in contrast to the original formulation with Shannon entropy, which scales variance with a fixed temperature parameter. Due to this property, our proposed algorithms can promote exploration when necessary, that is, the cost of the trajectory is high. The simulation results with two robotic systems with multimodal cost demonstrate the properties of the proposed algorithm.
format Preprint
id arxiv_https___arxiv_org_abs_2403_18130
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Generalized Maximum Entropy Differential Dynamic Programming
Aoyama, Yuichiro
Theodorou, Evangelos A.
Optimization and Control
Information Theory
34H05
We present a sampling-based trajectory optimization method derived from the maximum entropy formulation of Differential Dynamic Programming with Tsallis entropy. This method is a generalization of the legacy work with Shannon entropy, which leads to a Gaussian optimal control policy for exploration during optimization. With the Tsallis entropy, the policy takes the form of $q$-Gaussian, which further encourages exploration with its heavy-tailed shape. Moreover, the sampling variance is scaled according to the value function of the trajectory. This scaling mechanism is the unique property of the algorithm with Tsallis entropy in contrast to the original formulation with Shannon entropy, which scales variance with a fixed temperature parameter. Due to this property, our proposed algorithms can promote exploration when necessary, that is, the cost of the trajectory is high. The simulation results with two robotic systems with multimodal cost demonstrate the properties of the proposed algorithm.
title Generalized Maximum Entropy Differential Dynamic Programming
topic Optimization and Control
Information Theory
34H05
url https://arxiv.org/abs/2403.18130