AER1517 - Controls for robotics - Final project
Link for project report
Link for Demonstration video
Preferable environment: Ubuntu 16.04
To setup the simulator environment, please follow the simulator setup instructions here
After the simulator is setup, perform the following:
$ cd ~/aer1217/labs/src
Remove the existing aer1217_ardrone_simulator package, and replace it with the package in this repository
$ cd ~/aer1217/labs
$ catkin_make
Consult ROS Wiki if you encounter any issues setting up the package
To run FMPC, use the following command:
$ roslaunch aer1217_ardrone_simulator ardrone_simulator.launch
To run using the DNN output, change the flag (self.with_DNN) in MPC.py line 51 to True
(MPC.py is located under ~/aer1217/labs/src/aer1217_ardrone_simulator/scripts)
To record training data while running FMPC, run this command in separate terminal
$ rosbag record /aer1217/learning_state /aer1217/learning_input
To compile the training data:
$ cd ~/aer1217/labs/src/aer1217_ardrone_simulator/DNN
$ rosrun aer1217_ardrone_simulator compile_training_data.py
In a separate terminal
$ rosbag play <bag file recorded>
To train the DNN:
$ cd ~/aer1217/labs/src/aer1217_ardrone_simulator/DNN
$ python dnn_train.py
Course material from UTIAS AER1517 - Controls for Robotics
Course material and drone simulator from UTIAS AER1217 - Autonomous UAS course
Full references list can be found in the final report
Finally, special thanks for the authors of the following papers as it was influential for the work done in the project
Melissa Greeff and Angela P. Schoellig, Flatness-based Model Predictive Control for Quadrotor Trajectory Tracking, Proc. of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018, pp. 6740—6745
SiQi Zhou and Mohamed K. Helwa and Angela P. Schoellig, “Design of Deep Neural Networks as Add-on Blocks for Improving Impromptu Trajectory Tracking”, in Proc. of the IEEE Conference on Decision and Control (CDC), 2017, pp 5201—5207