Duckietown Shell

Duckietown Shell is a pure Python, easily distributable (no dependencies) utility for Duckietown.

The idea is that most of the functionality is implemented as Docker containers, and dt-shell provides a nice interface for that, so that user should not type a very long docker run command line.

Prerequisites

You must be using pip with Python 2.x to use Duckietown Shell:

$ pip --version # Should return something like: pip 18.0 from ...python2.7/site-packages/pip (python 2.7)

If not, first install virtualenv:

$ pip install virtualenv

Then create a new virtual environment, dts:

$ virtualenv -p <PATH_TO_PYTHON2_BINARY> dts

Then activate the dts environment:

$ source dts/bin/activate
(dts) $ 

Now, you now should be ready to install duckietown-shell. To deactivate dts later, run deactivate.

Installation

These installation steps make sure that you have a minimal "sane" environment, which includes:

1. Git and Git LFS;
2. Docker;
3. The Duckietown Shell.

Installation on Ubuntu 18.xx

Installs pip, git, git-lfs, docker, duckietown-shell:

$ sudo apt install -y python-pip git git-lfs

$ sudo apt install -y docker.io
$ sudo adduser `whoami` docker

$ sudo pip install --no-cache-dir -U duckietown-shell

Note: you need to log in and out to have the group change take effect.

Installation on Ubuntu 16.xx

Installs pip, git, git-lfs, docker, duckietown-shell:

$ sudo apt-get install software-properties-common  curl
$ sudo add-apt-repository ppa:git-core/ppa
$ curl -s https://packagecloud.io/install/repositories/github/git-lfs/script.deb.sh | sudo bash
$ sudo apt-get install  -y python-pip git git-lfs 

$ curl -fsSL https://get.docker.com | sudo bash
$ sudo usermod -aG docker `whoami` 

$ sudo pip install --no-cache-dir -U duckietown-shell

Note: you need to log in and out to have the group change take effect.

Installation in other operating systems

You will need to find the instructions for git, git-lfs, docker.

To install the shell, use:

$ sudo pip install --no-cache-dir -U duckietown-shell

The shell itself does not require any other dependency beside standard cross-platform Python libraries.

---

Commands for compiling the Duckumentation

To compile one of the books:

$ git clone https://github.com/duckietown/docs-duckumentation.git
$ cd docs-duckumentation
$ git submodule init
$ git submodule update
$ dts docs build

There is an incremental build system. To clean and run from scratch:

$ dts docs clean
$ dts docs build

TODO: link to docs

---

Duckietown Authentication Token setup

Run the command dts tok set to set the Duckietown authentication token:

$ dts tok set  

Instructions will guide you and you will be prompted for the token.

If you already know the token, then you can use:

$ dts tok set dt1-YOUR-TOKEN

Verifying that a token is valid

To verify that a token is valid, you can use:

$ dts tok verify dt1-TOKEN-TO-VERIFY

This exits with 0 if the token is valid, and writes on standard output the following json:

{"uid": 3, "expiration": "2018-09-23"}

which means that the user is identified as uid 3 until the given expiration date.

---

Commands for Duckiebot setup

Command for flashing SD card

This command will install DuckieOS on the SD-card:

$ dts init_sd_card

---

Command for starting ROS GUIs

This command will start the ROS GUI container:

$ dts start_gui_tools <DUCKIEBOT_NAME_GOES_HERE>

---

Command for calibrating the Duckiebot

This command will run the Duckiebot calibration procedure:

$ dts calibrate_duckiebot <DUCKIEBOT_NAME_GOES_HERE>

---

Commands for AI-DO 1

(TODO) AI-DO templates download

Not implemented yet

The subcommand get-template downloads the submission templates.

Downloads the current template:

$ dts aido1 get-template TASK-LANG
checking out repository...

Without arguments, the program writes a list of available templates.

(TODO) AI-DO submissions

Not implemented yet

The command submit submits the entry in the current directory:

$ dts aido1 submit

(TODO )Submissions status

The command status displays the status of the submitted entries:

$ dts aido1 status
jobname  task  docker hash  status
jobname  task  docker hash  status
...

---

(TODO) Commands for logs

Wrappers are provided for the EasyLogs commands.

$ dts logs summary
$ dts logs download
$ dts logs copy
$ dts logs details
$ dts logs make-thumbnails
$ dts logs make-video

Summary, details

Queries the list of logs that satisfy a query:

$ dts logs summary "vehicle:yaf,length:>120s"

Show more details about one log:

$ dts logs details 20171124170042_yaf

Download, copy

Downloads logs to the local computer:

$ dts logs download 20171124170042_yaf

Copies the logs to a specific directory:

$ dts logs copy -o ![output dir] 20171124170042_yaf
Creating ![output dir]/20171124170042_yaf.bag

Creating thumbnails

$ dts logs make-thumbnails 20171124170042_yaf

Creating videos:

Create a video of the camera data:

$ dts logs make-videos 20171124170042_yaf

Advanced configuration

By default, Duckietown Shell uses the folder ~/dt-data for storing log data and other cache data.

Alternatively, you can use the --dt-data argument to choose a different folder:

$ dts logs --dt-data ![dir] ![command]

Alternatively, the directory can be specified using the environment variable DT_DATA.

$ DT_DATA=/tmp/data dt logs summary

---

Information for Duckietown Shell developers

Docker

To launch the Duckietown Shell in Docker, run the following command:

$ docker run -it duckietown/duckietown-shell

Note: the Duckietown Shell is supposed to be run natively from the host.

Local commands development

Use the env variable to work on your local copy of the commands:

export DTSHELL_COMMANDS=/path/to/my/duckietown-shell-commands