Modify the function 1 in the video_control_1.0.py. Please use this file to test new output.
The diagram showing logics for Function 1 && Function 2. Diagram
key_code contains codes for controller, data collection, training, evaluation
- control_collectData
- Anafi_tracking.py: webcam localization system and localization publisher
- keyboard_ctrl.py: keyboard controller for Anafi
- video_control.py: collect reference path, collect control correction, data augmentation using active transfer learning
- NNtraining
- data_provider.py: read data from files and feed them to neural network
- models.py: the neural network architecture
- NNutils.py: helper functions
- train.py: main function for training the result
- evaluation.py: evaluate the network performance and draw plots
- webcam
- find_extrinsct.ipynb: find the extrinsct of the webcam for localization
Preparations:
sudo apt-get install rabbitmq-server
pip install -r requirements.txt
Starting the simulation:
sudo systemctl start firmwared
sphinx /opt/parrot-sphinx/usr/share/sphinx/drones/anafi4k.drone::stolen_interface=::simple_front_cam=true
Test the keyboard controller:
Start the Sphinx simulation
In another terminal, enter the olympe environment
cd key_code/control_collectData
python keyboard_ctrl.py