This file contains a sample demo of a time-invariant kalman filter at work. More...
Go to the source code of this file.
Functions | |
int | main (int argc, char *argv[]) |
This file contains a sample demo of a time-invariant kalman filter at work.
User needs to input a dynamical system to interact with kf A lds with the same params. as the kf is assumed here
User specified inputs are in the form of *.csv files contained in the same folder as this file main.cc. A Matlab m.file (test_kf.m) is provided to facilitate this process. See comments therein
Inputs: t_in: duration of expt. a_in, b_in, c_in, q_in, r_in, s_in: sys. params x_pred_0_in, p_pred_0_in: initializes kf y_pred_0_in, inno_cov_0_in: initalizes kf
Outputs: mainly for comparison with matlab code see test_kf.m These will be stored as *.csv files in the directory w_out, v_out: stochastic signals u_out, x_out, y_out: system signals x_pred_out, p_pred_end_out: 1-step ahead predictions & err. cov. x_hat_out, p_hat_end_out: filtered state estimates & err cov. y_pred_out, inno_cov_end_out: 1-step ahead mst. predictions & err cov. k_gain_end_out: kalman gain note that only the final err. covs are returned because k.f. converges
Definition in file kalman_main.cc.