FoodButler

Google Assistant Project for finding nearest grocery store with a given item.

Comment

FoodButler

Google Assistant Project for finding nearest grocery store with a given item. This project was done as part of the hackathon https://teens-take-on-covid.devpost.com/

About

This project uses Google Assistant as a user interface. This allows user to use normal English to shop for everday grocery items such as bread and milk. The google assistant then responds with the store which has the requested item and which is also closest to the user. This project uses LocationIQ for getting the geocoordinates of the stores. It gets the location of the user using permissions from Google Assistant. It then finds the store which is closest to the user. The information about the stores is stored in a Postgress database hosted by Heroku. The custom webhook used by Google Assistant for this project is also hosted on Heroku using Tomcat.

Components:

1. Dialogflow:

The exported Dialogflow file is in Other/FoodButler.zip You can import it into your Dialogflow projects. Defines the intents for:

  1. Listing the stores (List)
  2. Getting the closest store given an item (user_location and FindFood)
  3. Welcome (Default Welcome Intent) and Goodbye intent (End)

Sample invocations:

  1. Find me milk
  2. Find me bread
  3. List
  4. Thank you
  5. done

2. LocationIQ

This project uses a free location APIs from LocationIQ. You can set uo your account and change the key in the HTTPClient.java file Forward Geocoding API was used to get the geocoordinates of the supported stores.

3. Postgress Datatbase

The Postgress Database was used to store detailes of the supported grocery stores. It currently only has a handfull of local grocery stores. More can be added to the stores table. The items supported sucha as bread, milk are in the stock table. The sql dump is in Other/sqlfile.sql.

4. Webhook

The webhook is coded in Java and it is hosted on Heroku using tomcat. Maven is used to assemble the webapp. The GoogleWebhook.java gets the call from Google Assistant and then it access the databuse using the DBHelper.java and gets the geocoordinates from HTTPClient.java and then calculates the distance using the DistanceCalculator.java.

Built With

Share this project:

Updates