(PyARdrone + Tensorflow object recognition)

Be sure to have already installed Python’s package manager

sudo apt install python3-pip

Install the following packages:

sudo pip3 install pyardrone
sudo pip3 install numpy
sudo pip3 install tensorflow
sudo pip3 install Cython
sudo pip3 install pillow
sudo pip3 install lxml
sudo pip3 install jupyter
sudo pip3 install matplotlib

Download or clone tensorflow models repository from here:

https://github.com/tensorflow/models

Extract the zip file if downloaded.

If you don’t have protobuf already installed, download the latest version for your corresponding python version from here:

https://github.com/google/protobuf/releases

Extract the zip file and go to the protobuf folder to execute the following commands:

sudo ./configure
sudo make check
sudo make install

If you get the following error:

"protoc: error while loading shared libraries: libprotoc.so.15: cannot open shared object file: No such file or directory"

Just execute the following command:

sudo ldconfig

If you already had/have profobuf then go to “tensorflow/models/research/” from the downloaded repo folder and launch the following commands:

protoc object_detection/protos/*.proto --python_out=.
export PYTHONPATH=$PYTHONPATH:`pwd`:`pwd`/slim

You’re ready to start using tensorflow. For more information:

https://www.tensorflow.org/install/install_linux
https://github.com/tensorflow/models/tree/master/research/object_detection

After correctly installing tensorflow and all the libraries, go to the object detection folder, "models-master/research/object_detection" and create your new file in this path, or simply download the code file from:

https://github.com/diegoballest/ardrone-tec

Copy the "drone.py" file to the path and run it with the following command:

sudo python3 drone.py

Make sure that your first execution is in a local wifi network, this will download the model file to make the object detection work.

After the first execution, connect to the drones network and also you can comment the following lines from the "drone.py" file:

opener = urllib.request.URLopener()
opener.retrieve(DOWNLOAD_BASE + MODEL_FILE, MODEL_FILE)

This is because every time you execute the file, this tries to download the model file and takes more time to execute it.