Abstract:Continuous-time (CT) modeling has proven to provide improved sample efficiency and interpretability in learning the dynamical behavior of physical systems compared to discrete-time (DT) models. However, even with numerous recent developments, the CT nonlinear state-space (NL-SS) model identification problem remains to be solved in full, considering common experimental aspects such as the presence of external inputs, measurement noise, latent states, and general robustness. This paper presents a novel estimation method that addresses all these aspects and that can obtain state-of-the-art results on multiple benchmarks with compact fully connected neural networks capturing the CT dynamics. The proposed estimation method called the subspace encoder approach (SUBNET) ascertains these results by efficiently approximating the complete simulation loss by evaluating short simulations on subsections of the data, by using an encoder function to estimate the initial state for each subsection and a novel state-derivative normalization to ensure stability and good numerical conditioning of the training process. We prove that the use of subsections increases cost function smoothness together with the necessary requirements for the existence of the encoder function and we show that the proposed state-derivative normalization is essential for reliable estimation of CT NL-SS models.

Subspace Identification of Continuous-Time Models Using Generalized Orthonormal Bases

Subspace Identification for Continuous-Time Errors-in-variables Model from Sampled Data

Closed-loop Subspace Identification for Continuous-time Models

A new closed-loop identification technique for step-response model based on subspace approach

A New Subspace Identification Approach Based on Principal Component Analysis and Noise Estimation

Closed-loop continuous-time Laguerre-based subspace identification via orthogonal projection approach

Recursive Subspace Identification of Continuous-Time Systems Using Generalized Poisson Moment Functionals

Subspace Model Identification for Industrial Processes

Continuous-discrete Extended Kalman Filter Based Parameter Identification Method for Space Robots in Postcapture

Improvement of subspace identification

A Dual-Frequency Approach Based on the Subspace Method for Identifying Nonlinear Vibration Structures

Closed-loop Delta-Operator-based Subspace Identification for Continuous-Time Systems Utilising the Parity Space.

A New Excitation Scheme for Closed-Loop Subspace Identification Using Additional Sampling Outputs and Its Extension to Instrumental Variable Method.

Subspace System Identification Methods for Structural Modal Parameters

Subspace-Based Identification of Linear Time-Varying System

System Identification Based on Invariant Subspace

An Improved Subspace Identification Method with Variance Minimization and Input Design

An Improved Consistent Subspace Identification Method Using Parity Space for State-space Models

Identification of hybrid linear time-invariant systems via subspace embedding and segmentation (SES)

Continuous-time identification of dynamic state-space models by deep subspace encoding

Subspace tracking for online system identification