Estimation in non-linear state-space models.

Robust estimator for non-linear state-space models with state-dependent noise.

The Robust Non-Linear Estimator (RNLE) is a robust estimator for non-linear state-space models with state-dependent noise. It takes a sequence of input-output pairs and estimates the corresponding sequence of states. The estimates are found by solving an iteratively-reweighted non-linear least-squares problem. The solver is robust to outliers and accepts missing values.

This submission includes an initialization script, a test function and a technical report. The initialization script adds all relevant directories and sub-directories to the MatLab path and compiles two MEx files, both of them necessary for the code to run. The test function creates a short animation (in AVI format) showing how the state sequence is estimated from extremely noisy data. The technical report contains a detailed derivation of the theory behind the code.

If you find this submission useful for your research/work please cite my technical report and/or my MathWorks community profile. Feel free to contact me directly if you have any technical or application-related questions.

INSTRUCTIONS:

After downloading this submission, extract the compressed file inside your MatLab working directory and run the initialization script (init.m). Then, run the test function (TestRNLE.m) for a demonstration.