Learning Based Observer

Learning-based observers are computational methods that estimate the unobservable internal states of dynamic systems, surpassing traditional model-based approaches in handling nonlinearities and uncertainties. Current research focuses on leveraging neural networks, including neural ODEs and Fourier Neural Operators, and meta-learning techniques to improve accuracy, robustness, and computational efficiency, particularly for nonlinear systems and partial differential equations. These advancements are significant for various applications, such as improved control systems, anomaly detection in sensor networks, and more efficient state estimation in robotics and other fields.

Papers

June 7, 2024

Creating an AI Observer: Generative Semantic Workspaces
Pavan Holur, Shreyas Rajesh, David Chong, Vwani Roychowdhury
BERT Model Semantic Map Semantic Space Working Memory Learning Based Observer

October 30, 2023

Adaptive Meta-Learning-Based KKL Observer Design for Nonlinear Dynamical Systems
Lukas Trommer, Halil Yigit Oksuz
Adaptive Importance Nonlinear Dynamic State Space Observer Design Learning Based Observer

March 31, 2023

Learning-based Observer Evaluated on the Kinematic Bicycle Model
Agapius Bou Ghosn, Philip Polack, Arnaud de La Fortelle
Kalman Filter Extended Kalman Filter Observer Design Learning Based Observer

December 1, 2022

Learning Robust State Observers using Neural ODEs (longer version)
Keyan Miao, Konstantinos Gatsis
Neural ODE Learning Based Observer Unknown Nonlinear Dynamic Luenberger Observer

November 28, 2022

Machine Learning Accelerated PDE Backstepping Observers
Yuanyuan Shi, Zongyi Li, Huan Yu, Drew Steeves, Anima Anandkumar, Miroslav Krstic
State Estimation Learning Based Observer PDE Backstepping Prescribed Time

August 27, 2022

Neural Observer with Lyapunov Stability Guarantee for Uncertain Nonlinear Systems
Song Chen, Shengze Cai, Tehuan Chen, Chao Xu, Jian Chu
Lyapunov Function Uncertain Dynamic Disturbance Rejection Observer Design Learning Based Observer

August 26, 2022

Learning From Positive and Unlabeled Data Using Observer-GAN
Omar Zamzam, Haleh Akrami, Richard Leahy
LeArning Abstract Generative Adversarial Network Unlabeled Data GAN Training Positive Emotional Language GAN Algorithm GAN Discriminator Learning Based Observer

April 4, 2022

Distributed Anomaly Detection and Estimation over Sensor Networks: Observational-Equivalence and Q-Redundant Observer Design
Mohammadreza Doostmohammadian, Themistoklis Charalambous
Anomaly Detection Estimation Task Sensor Network Linear Dependence Model Equivalence Residual Information Optimal Sensor Learning Based Observer

February 1, 2022

Adversarial Imitation Learning from Video using a State Observer
Haresh Karnan, Garrett Warnell, Faraz Torabi, Peter Stone
Source Video Adversarial Imitation Learning Adversarial Imitation State Observation Learning Based Observer Generative Adversarial Imitation Generative Visual