State Estimation and Nonlinear System Modeling Advances

The field of nonlinear system modeling and state estimation is witnessing significant developments, with a focus on improving accuracy, robustness, and interpretability. Researchers are exploring innovative approaches to integrate system dynamics and sensor data into physics-informed learning processes, enabling more accurate state estimation and better handling of noisy measurements. Noteworthy papers include PINN-Obs, which introduces a novel Adaptive Physics-Informed Neural Network-based Observer for accurate state estimation in nonlinear systems. Noisy PDE Training Requires Bigger PINNs proves a lower bound on the size of neural networks required for effective PINN training with noisy data. These advances have the potential to enhance estimation accuracy and robustness in various applications, including control systems and state-space modeling.

Sources

PINN-Obs: Physics-Informed Neural Network-Based Observer for Nonlinear Dynamical Systems

Noisy PDE Training Requires Bigger PINNs

Perspective Chapter: Insights from Kalman Filtering with Correlated Noises Recursive Least-Square Algorithm for State and Parameter Estimation

Space-Filling Regularization for Robust and Interpretable Nonlinear State Space Models

An Empirical Bernstein Inequality for Dependent Data in Hilbert Spaces and Applications

Built with on top of