main.m

clear all; close all; clc;

% Define path to a urdf file
path_to_urdf = 'ur10e.urdf';


% Generate functions for dynamics based on Lagrange method
% Note that it might take some time
% generate_rb_dynamics(path_to_urdf);
generate_friction_eq();


% Generate regressors for inverse dynamics of the robot, friction and load
% Note that it might take some time
% generate_rb_regressor(path_to_urdf);
% generate_load_regressor(path_to_urdf);


% Run tests
test_rb_inverse_dynamics()
test_base_params()


% Perform QR decompostion in order to get base parameters of the robot
include_motor_dynamics = 1;
[pi_lgr_base, baseQR] = base_params_qr(include_motor_dynamics);


% Estimate drive gains
drive_gains = estimate_drive_gains(baseQR, 'PC-OLS');
% Or use those found in the paper by De Luca
% drive_gains = [14.87; 13.26; 11.13; 10.62; 11.03; 11.47]; 


% Estimate dynamic parameters
path_to_est_data = 'ur-20_02_10-30sec_12harm.csv';      idxs = [635, 3510];
% path_to_data = 'ur-20_02_12-40sec_12harm.csv';    idxs = [500, 4460];    
% path_to_data = 'ur-20_02_05-20sec_8harm.csv';     idxs = [320, 2310];
% path_to_data = 'ur-20_02_12-50sec_12harm.csv';    idxs = [355, 5090];
sol = estimate_dynamic_params(path_to_est_data, idxs, ...
                              drive_gains, baseQR, 'PC-OLS');

                          
% Validate estimated parameters
path_to_val_data = 'ur-20_01_17-ptp_10_points.csv';     idxs = [700, 4200];

rre = validate_dynamic_params(path_to_val_data, idxs, ...
                              drive_gains, baseQR, sol.pi_b, sol.pi_fr)