可能近似正确的非线性模型预测控制(PAC-NMPC)
机器人学
2023-09-14 v3
摘要
随机非线性模型预测控制(SNMPC)的方法通常对系统动力学做出限制性假设,并依赖近似来刻画底层不确定性分布的演化。因此,它们往往无法捕捉更复杂的分布(例如非高斯或多模态),也不能提供准确的性能保证。本文提出一种基于采样的 SNMPC 方法,其利用最近导出的样本复杂度界来认证反馈策略的性能,而无需对系统动力学或底层不确定性分布做出假设。通过并行化我们的方法,我们能够在仿真以及硬件上利用 1/10 比例拉力车和 24 英寸翼展固定翼无人机(UAV)演示具有统计安全保证的实时滚动时域 SNMPC。
引用
@article{arxiv.2210.08092,
title = {Probably Approximately Correct Nonlinear Model Predictive Control (PAC-NMPC)},
author = {Adam Polevoy and Marin Kobilarov and Joseph Moore},
journal= {arXiv preprint arXiv:2210.08092},
year = {2023}
}
备注
2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works