Neural Network Observer-Identifier for MIMO Nonlinear Systems

Implementation of a recurrent neural network based Observer/Identifier for highly Non-Linear Dynamical Systems, using the Lorenz Attractor as a test case.

Note: In this context, an observer uses system output as input to estimate states, whereas an identifier uses system states as input to estimate unknown dynamics.

---

References

• Talebi, H.A., et al. "Neural Network-Based State Estimation of Nonlinear Systems." Lecture Notes in Control and Information Sciences 395 (2010): DOI. Springer Science+Business Media, LLC.