Saved in:
Bibliographic Details
Main Authors: Walker, Markus, Frisch, Daniel, Hanebeck, Uwe D.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.05706
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910206567383040
author Walker, Markus
Frisch, Daniel
Hanebeck, Uwe D.
author_facet Walker, Markus
Frisch, Daniel
Hanebeck, Uwe D.
contents Cross-entropy method model predictive control (CEM--MPC) is a powerful gradient-free technique for nonlinear optimal control, but its performance is often limited by the reliance on random sampling. This conventional approach can lead to inefficient exploration of the solution space and non-smooth control inputs, requiring a large number of samples to achieve satisfactory results. To address these limitations, we propose deterministic sampling CEM (dsCEM), a novel framework that replaces the random sampling step with deterministic samples derived from localized cumulative distributions (LCDs). Our approach introduces modular schemes to generate and adapt these sample sets, incorporating temporal correlations to ensure smooth control trajectories. This method can be used as a drop-in replacement for the sampling step in existing CEM-based controllers. Experimental evaluations on two nonlinear control tasks demonstrate that dsCEM consistently outperforms state-of-the-art iCEM in terms of cumulative cost and control input smoothness, particularly in the critical low-sample regime.
format Preprint
id arxiv_https___arxiv_org_abs_2510_05706
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Sample-Efficient and Smooth Cross-Entropy Method Model Predictive Control Using Deterministic Samples
Walker, Markus
Frisch, Daniel
Hanebeck, Uwe D.
Systems and Control
Cross-entropy method model predictive control (CEM--MPC) is a powerful gradient-free technique for nonlinear optimal control, but its performance is often limited by the reliance on random sampling. This conventional approach can lead to inefficient exploration of the solution space and non-smooth control inputs, requiring a large number of samples to achieve satisfactory results. To address these limitations, we propose deterministic sampling CEM (dsCEM), a novel framework that replaces the random sampling step with deterministic samples derived from localized cumulative distributions (LCDs). Our approach introduces modular schemes to generate and adapt these sample sets, incorporating temporal correlations to ensure smooth control trajectories. This method can be used as a drop-in replacement for the sampling step in existing CEM-based controllers. Experimental evaluations on two nonlinear control tasks demonstrate that dsCEM consistently outperforms state-of-the-art iCEM in terms of cumulative cost and control input smoothness, particularly in the critical low-sample regime.
title Sample-Efficient and Smooth Cross-Entropy Method Model Predictive Control Using Deterministic Samples
topic Systems and Control
url https://arxiv.org/abs/2510.05706