Lumos Maxima 🍻 entertainment gadget + Alexa

TL;DR;

Play games with connected cups and Alexa.

Alexa + 🍻 = 🎉

We wanted to see if we could introduce Alexa into social occasions such as drinking and games. Lumos Maxima attempts to do this by hooking up some cups with a few lights and ESP8266 Wi-Fi microchips.

How it works:

Some Uses Cases:

• Parties

• Night clubs

• Team games

• Game nights

• Assign cups (never lose your cup again) e.g. "Alexa, where is Magic Mike's cup"

Why with an App?

The app introduces some fun custom features such as allowing you to name each cup (opening up a lot of fun possibilities and games) and controlling each cup individually. It also allows you to play the games without Alexa.

Taking it further

We plan to build out the games and continue to develop the prototype. Currently on our todo list:

• Add battery back to each node (currently we power it via USB)

• 3D print base that can attach to various cups (replace the cardboard/plastic bags)

• Replace the plastic cups for some nice glass tumblers or mugs

• Add support for additional bot platforms (slack useful during work etc.)

• Trivia questions with score keeping

• Name games using the nicknames assigned from the app

• Split people into teams

• VOS games

• Sentence forming games

• Add buttons or other sensors to cups to enhance gameplay

Try it for yourself

The Alexa Skill is currently being published. We will add a link here once its ready. In the mean time you can build it yourself using the source code. There are three parts to the project. 1) VUI 2) Native App and 3) Arduino Hardware. Checkout the README in each folder for instructions on how to get setup.

---Building The Cups---

Each cup runs as a node and defaults to being a "Slave". This means that they can receive commands or pass them on but cannot give commands. The cup that runs as the "Master" node can give the commands and is connected directly to the controller via the `RX` and `TX` pins on the boards (see below). The wiring to the LED and node is identical for all cups.

The cups communicate via a private wireless mesh network that the nodes make. This allows for near instant communication allowing us to create some cool light syncing effects.

The controller connects to the local wifi (setup via the app) which allows it to then connect directly to Firebase. The controller then waits for commands and also sends updates of the cups current status. The commands can be sent via Alexa or the App.

---Building The VUI---

Developed using the following stack:

• Dialogflow

• Firebase

• Firebase function + AWS Lambda API Gateway and Alexa Skills Set function

Setting up

Assuming you have Dialogflow, Firebase, and or AWS account access

Dialog Flow

I. Create DIALOGFLOW AGENT

• Create new Agent in DialogFlow from the console

• You can either import the intents 1 by 1 or zip assets/dialogflow folder and import it.

• to import, cleck on gear icon on right panel next to the agent name

• click on Export and Import tab

• click on IMPORT FROM ZIP button, and upload the zip file

II. Setting up Dialogflow Agent webhook fulfillment - click on Fulfillment menu on the left side of your console - toggle enabled switch to enable webhook - enter the URL of webhook end point - fill up the necessary authentication detail for the webhook access, for AWS Lambda webhook, add x-api-key as one of the header, and grab API key from AWS API Gateway console - select Enable webhook for all domains for DOMAINS - P.S: 2 code base options available, can either use Firebase functions or AWS Lambda API Gateway

Firebase

I. Importing Data

• go to Realtime Database console

• click on menu icon on the right side (three dots icon), click on import JSON

• JSON file is located at assets/firebase/realtime-db-data.json

II. Deploying Firebase function (should you choose Firebase function as your agent webhook fulfillment)

• Pre-requisite

• install Firebase CLI npm install -g firebase-tools

• once done, login to your firebase account by running firebase login in your terminal

• Deployment

• go to firebase-webhook/.firebaserc and key in your firebase project ID and save the file

• open your terminal and go to firebase-webhook/functions, and do the following:

• run npm install

• run firebase deploy

• take note that on first deployment, it will take longer time than the subsequent deployment.

• once the deployment is done, grab the firebase function url, and use that url as the Dialogflow fulfillment webhook URL

AWS Lambda

I. Deploying AWS Lambda API Gateway function - Dialogflow webhook (should you choose AWS Lambda API Gateway function as your agent webhook fulfillment)

• AWS Lambda Console

• go to AWS Lambda console

• create a new function

• Fill up all the function details i.e. name, role.

• choose Node.js 6.10 as Runtime

• choose Author from scratch

• click on Create function button

• it should bring you to the next step, add API Gateway triggers (in Designer section, left panel, there is a list of available triggers)

• Next, you will need to configure the triggers:

• Enter your api name, e.g. test-api

• Choose Open with access key as your Security preference

• Click on Add buttons

• Save the function

• it should generate an invoke url, use that url as the Dialogflow fulfillment webhook URL

• Now for the code deployment:

• on your machine, go to aws-lambda-webhook folder

• create a folder called keys, put the firebase service key file and name it firebase-service-key.json + create a file called database-details.js and add in the following content

• module.exports = { databaseURL: '[FIREBASE database URL]'};

• once done, run npm install

• run npm run zip

• it will create a zip folder called aws-lambda-webhook.zip

• back to the AWS Lambda console, on the function settings page, under Designer section, click on the function name button from the graph

• under Function code section > Code entry type, select Upload a .ZIP file

• upload the zip file

• save the function

• it will take a while to deploy

• API Gateway Console

• look for the api name e.g. test-api, and click on resources

• under actions drop down menu, create method for POST request

• set lambda region as us-east-1 + enter the lambda function name associated with this API

• click on save

• it will load POST method dashboard, go to Method Request

• set API Key Required to true

• add in new headers called x-api-key and set is as required

• follow the instructions from this blog to create request object mapping

• once done, create API Keys, from left menu panel, click 'API Keys'

• under actions drop down menu, create API Key

• go to Usage plans, from left menu panel, click 'Usage plans'

• create new usage plan

• enter usage plan name

• set the rest of access limit related stuff and save

• from the usage plans dashboard, add API stage.

• once done, go to API Keys tab and add API key to the usage plan

II. Alexa skills

• Amazon Developer console - Creating custom Alexa skills

• go to Amazon Developer Console and login

• click on Alexa tab > click on Get started on alexa skills > Add a new skill

• enter skill information details, and go to next step (Interaction Model)

• enter both Intent schema + Sample utterance by referring to the assets file provided at alexa-skills/assets

• under Configuration steps, fill up the AWS Lambda ARN details

• and... done!

• AWS Lambda Console

• go to AWS Lambda console, make sure to set the region to US East (N.Virginia)

• create a new function

• Fill up all the function details i.e. name, role.

• choose Node.js 6.10 as Runtime

• choose Author from scratch

• click on Create function button

• it should bring you to the next step, add Alexa Skills Kit triggers (in Designer section, left panel, there is a list of available triggers)

• Next, you will need to configure the triggers:

• enter alexa skill ID (grab it from Amazon Developer Console > alexa skills)

• save it

• Now for the code deployment:

• on your machine, go to alexa-skills folder

• create a folder called keys, put the create a file called alexa-dialogflow-keys.js and add in the following content

• module.exports = { alexaAppId: '[alexa APP ID]', apiAIDeveloperAccessToken: '[dialogflow developer access token]'};

• once done, run npm install

• run npm run zip

• it will create a zip folder called alexa-skills.zip

• back to the AWS Lambda console, on the function settings page, under Designer section, click on the function name button from the graph

• under Function code section > Code entry type, select Upload a .ZIP file

• upload the zip file

• save the function

• it will take a while to deploy

• ready to test!

---Setting Up The App---

Standard compliant React Native App Utilizing Ignite

Step 1: git clone this repo:

Step 2: cd to the cloned repo:

Step 3: Install the Application with yarn or npm i

How to Run App

• cd to the repo

• Run Build for either OS

• for iOS

• run react-native run-ios

• for Android

• Run Genymotion

• run react-native run-android

Secrets

This project uses react-native-config to expose config variables to your javascript code in React Native. You can store API keys and other sensitive information in a .env file:

API_URL=https://myapi.comGOOGLE_MAPS_API_KEY=abcdefgh

and access them from React Native like so:

import Secrets from 'react-native-config'Secrets.API_URL  // 'https://myapi.com'Secrets.GOOGLE_MAPS_API_KEY  // 'abcdefgh'

The .env file is ignored by git keeping those secrets out of your repo.

Get started:

• Copy .env.example to .env

• Add your config variables

• Follow instructions at https://github.com/luggit/react-native-config#setup

• Done!