Aerodynamic Model Identification using the Two-Step Approach
Aircraft aerodynamic model identification using flight test data using the two-step approach of system identification has been performed.
Main script to be run : main.m Refer to the instructions mentioned in the comments of the above file.
File list :
Data Files :
simdata2018 > da3211.m, dadoublet.m, de3211.m, dr3211.m, drdoublet.m Data files that were provided with the assignment. Contain flight test data for various manoeuvres.
Functions:
aero_fm.m : Computes aerodynamic forces and moments
calc_f.m : Defines state transition functions
calc_G.m : Computes process noise coefficient matrix
calc_h.m : Defines observation functions
data_preprocessing.m : Pre-processes flight data
getposition.m : Compute aircraft positions from aircraft and win velocities
integration_rk4.m : Fourth order Runge-Kutta integrator
jacob.m : Computes Jacobian of the input function
kalman_filter.m : Extended Kalman Filter (EKF) implementation
LR_states.m : Compute state variables of the linear regression model
param_est.m : Ordinary Least Squares parameter estimator
param_est_alternate.m : Parameter estimator for the alternate model structure
param_val.m : Validation of the computed parameter values
param_val_alternate.m : Validation of computed parameter values for the alternate model structure
Scripts :
generate_noise_vectors.m : Generate sensor noise vectors
main.m : Main script
params.m : Import aircraft parameters