Perform system identification to estimate desired parameters



MCMC Markov Chain Monte Carlo.
matrix/linfitsvd Linear Parameter Estimation with Singular Value Decomposition


In this section we perform system identification to estimate desired parameters. We follow two approaches. The Marcov Chain Monte Carlo and the Linear Parameter Estimation with Singular Value Decomposition. For both approaches, some definitions are needed. First we load all the objects created for this simulation.

    
clear 

%% Load Objects (signals and model)
in    = matrix('input.mat');
out   = matrix('output.mat');
noise = matrix('noise.mat');
mdl   = ssm('fitting_model.mat');

Again we will have to construct cell arrays that will contain the names of the inputs and the outputs of the system.
 
  
% Input & Output names 
inNames  = { 'GUIDANCE.IFO_x1'  'GUIDANCE.IFO_x12'};
outNames = {'DELAY_IFO.x1' 'DELAY_IFO.x12'};

The last common action for both methods is to again define the parameters to estimate and also the initial values of the parameters.
 
  
% Define the parameters to estimate and the starting values. 
params = {'FEEPS_XX','CAPACT_TM2_XX','IFO_X12X1','EOM_TM1_STIFF_XX','EOM_TM2_STIFF_XX'};
values = [1 1 0.0001 1.e-6 1.e-6];
 

In the following sections, the two approaches to system identification are presented:

A) Markov Chain Monte Carlo

B) Linear Parameter Estimation with Singular Value Decomposition



©LTP Team