发表机构
Systems and Control Laboratory, HUN-REN Institute for Computer Science and Control; Control Systems Group, Eindhoven University of Technology（系统与控制实验室，HUN-REN计算机科学与控制研究所; 控制系统组，埃因霍温理工大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究基于深度学习的非线性系统识别中神经状态空间模型的在线学习，提出批处理学习管道和直接递归识别算法，经收敛性分析和仿真验证，该方法能实现高效在线自适应且模型精度高。
AI 中文摘要
基于深度学习的非线性系统识别的最新进展，使得基于编码器的神经状态空间（ANN-SS）模型估计得以实现，该模型在离线设置中通过从过去的输入输出数据估计初始模型状态，取得了最优性能。这些方法通常用于基于多步射击的离线识别，而这些模型的在线学习在很大程度上仍未得到探索。本文提出了一种基于子空间编码器的ANN-SS模型的批处理学习管道和直接递归识别算法。我们对递归公式进行了收敛性分析，并通过大量的仿真研究验证了其性能。结果表明，该方法能够实现高效的在线自适应，且模型精度高。
英文摘要
Recent advances in deep-learning-based nonlinear system identification have led to encoder-based estimation of neural state-space (ANN-SS) models that achieve state-of-the-art performance in offline settings by estimating initial model states from past input-output data. These methods are typically used in multiple-shooting-based offline identification, and online learning of these models remains largely unexplored. This paper presents a batch-wise learning pipeline and a direct recursive identification algorithm for subspace encoder-based ANN-SS models. We provide convergence analysis of the recursive formulation and validate its performance through extensive simulation studies. The results demonstrate that the proposed approach enables computationally efficient online adaptation with high model accuracy.
CommentsSubmitted to L-CSS. Extended version
