build details

Show: section status errors & todos local changes recent changes last change in-page changes feedback controls

Demo

Modified 2020-06-24 by Luzian Bieri

This section presents a demo of the hardware benchmark where all is run on your own machine.

Start the API

Modified 2020-07-20 by Luzian Bieri

The DT_APP_SECRET and APP_ID can be retrieved by either asking via the Slack or by using the Web-Debugger in the Duckietown Diagnostics within a (GET-)request header.

Saving data locally

Modified 2022-11-14 by Andrea Censi

In order to start the API which saves data on your local machine use the following command:

laptop $ docker run -v /PATH/TO/DATA/DIR/:/data/ -it -e LOCAL=True -p 5000:5000 -e APP_SECRET=DT_APP_SECRET -e APP_ID=APP_ID --rm duckietown/dt-hardware-benchmark-backend:daffy

Saving data online

Modified 2022-11-14 by Andrea Censi

Careful the API address in the frontend is hardcoded and thus needs to be adjusted

In order to start the API which saves data on S3 and a MySQL database use the following command:

laptop $ docker run -dit -p 5000:5000 -e MYSQL_USER=DB_USER -e MYSQL_PW=DB_PW -e MYSQL_URL=DB_URL -e MYSQL_DB=DB_NAME -e AWS_SECRET_ACCESS_KEY=AWS_SECRET_ACCESS_KEY -e AWS_ACCESS_KEY_ID=AWS_ACCESS_KEY_ID -e APP_SECRET=DT_APP_SECRET -e APP_ID=APP_ID--rm duckietown/dt-hardware-benchmark-backend:daffy

Start the Frontend

Modified 2022-11-14 by Andrea Censi

Adjust the API address if wanted to run online

laptop $ docker run -it -p 3000:80 --rm duckietown/dt-hardware-benchmark-frontend:daffy

It then is reachable under localhost:3000

Run the benchmark

Modified 2020-07-24 by Luzian Bieri

Ensure that Lane following runs on your duckiebot, if necessary consult the resp. documentation.

You need two terminals, another one to prepare the start of lane following as described in the output of the CLI. Plug in a USB Stick int the top left port of the Duckiebot.

Now it is time to run the benchmark you can use the dts:

laptop $ dts benchmark BOTNAME -a YOUR_LOCAL_IP_ADDRESS:5000

If your API is running online, enter the API_URL including the protocol (i.e. https://).