System Identification, Modeling, and Control of Linear and Nonlinear Systems
Currently working on controller and online modeling.
***NOTE: *** The files below needs to be revised and are currently inside Archive.
predict_gtm.py:
Loads previously collected data from NASA GTM Matlab sim and uses deep learning to predict model dynamics. It uses 4 states (beta, p, q, phi) and 2 controls (?).
main.py:
This file loads a previously trained model (saved inside 'models' folder) and a tries to control the plant using a modified version of iLQR controller.
predict_gtm_full.py:
Loads previously collected data from NASA GTM Matlab sim and uses deep learning to predict model dynamics. It uses 12 states (beta, p, q, phi) and (?) controls (?).
All the packages and dependencies are being managed using Anaconda3. Please go ahead and install Anaconda3 (Python 3.5 or greater).
Clone this repository. Go to its main folder and run:
./setup_linux.sh
Everything should be installed by now. If there's any problem during the installation, please contact Vinicius Goecks at viniciusguigo@gmail.com.
Before using SysId_L2Fly, activate the environment:
source activate sysid_l2fly
You might want to deactivate it eventually:
source deactivate
Jack Han-Hsun Lu: ineffable1201@gmail.com
Vinicius G. Goecks: viniciusguigo@gmail.com