FCND-Term1-Starter-Kit

Starter kit for FCND Term 1. Python 3 is used for the entirety of term 1.

Packages used:

matplotlib
jupyter
udacidrone. Due to udacidrone being updated as the ND progresses, it's recommended for new projects you update udacidrone with pip install --upgrade udacidrone.
visdom

The recommended way to get up and running:

Anaconda Environment

Get started here. More info here.

Supported Sytems: MacOS, Windows, Linux

Troubleshooting

NOTE: If future is not installed prior to pymavlink this will output an error message similar to the following:

  # omitted Traceback
  ModuleNotFoundError: No module named 'future'
  
  ----------------------------------------
  Failed building wheel for pymavlink
  Running setup.py clean for pymavlink
Failed to build pymavlink
Installing collected packages: torchfile, pillow, idna, chardet, urllib3, requests, visdom, future, lxml, pymavlink, utm, websockets, uvloop, udacidrone
  Running setup.py install for pymavlink ... done
  Running setup.py install for udacidrone ... done
Successfully installed chardet-3.0.4 future-0.16.0 idna-2.6 lxml-4.1.1 pillow-5.0.0 pymavlink-2.2.8 requests-2.18.4 torchfile-0.1.0 udacidrone-0.1.0 urllib3-1.22 utm-0.4.2 uvloop-0.9.1 visdom-0.1.7 websockets-4.0.1

If the output is Successfully installed ... pymavlink .. it's likely the install is fine. You can check by running:

python -c "import pymavlink"

if the install was successful there should be no output.

NOTE: If you're on MacOS you'll need to install developer tools to install pymavlink. You can do this by typing the following into a shell:

xcode-select --install

NOTE: On Windows you may need to open a terminal/powershell as an administrator. This can be done by right-clicking the program and selecting "Run as Administrator".

Docker Conda Environment

I created a dockerfile to create docker image with Ubuntu 18.04 and miniconda 4.3.11 to install and create the necessary conda environment.

The docker image can be built using following command -
```
cd docker/
docker build -t <my-image-name> .
```
The docker container can be started using following command -
```
docker run --network host -it <my-image-name>
```
You can mount the a directory from your pc to the docker container to access the code.
```
docker run --network host -it -v /path/on/host:/path/in/container <my-image-name>
```
You can use the -v option multiple times to mount multiple paths from your local PC to a single container. Here is an example command to mount two directories:
```
docker run --network host -it -v /path/to/local/dir1:/mount/path/dir1 -v /path/to/local/dir2:/mount/path/dir2 my-image-name
```
Once inside the container source the environment
```
source activate fcnd_new
```

To run jupyter notebook in the docker container -

jupyter notebook --ip=0.0.0.0 --port=8888 --allow-root

Solutions for MacOS with Apple Silicon (e.g. M1)

I created a dockerfile to create docker image with Ubuntu 18.04 and miniconda 4.3.11 to install and create the necessary conda environment.

The docker image can be built using following command -

cd docker/
docker build --platform linux/x86_64 -t <my-image-name> .

The docker container can be started using following command -

docker run -p 0.0.0.0:8888:8888/tcp --platform linux/x86_64 -it <my-image-name>

You can mount the a directory from your pc to the docker container to access the code.

docker run -p 0.0.0.0:8888:8888/tcp --platform linux/x86_64 -it -v /path/on/host:/path/in/container <my-image-name>

You can use the -v option multiple times to mount multiple paths from your local PC to a single container. Here is an example command to mount two directories:

docker run -p 0.0.0.0:8888:8888/tcp --platform linux/x86_64 -it -v /path/to/local/dir1:/mount/path/dir1 -v /path/to/local/dir2:/mount/path/dir2 my-image-name

Once inside the container source the environment
```
source activate fcnd_new
```

To run jupyter notebook in the docker container -

jupyter notebook --ip=0.0.0.0 --port=8888 --allow-root

Using ipython in docker container to control the drone

Use the host and add additional ports to the container for controlling the udacidrone simulator

docker run --add-host=host.docker.internal:host-gateway -p 8888:8888 -p 5760:5760  --platform linux/x86_64 -it -v /path/to/local/dir1:/mount/path/dir1 -v /path/to/local/dir2:/mount/path/dir2 my-image-name

Start ipython first and controll the udacidrone simulator with changing the connection argument