RIDEMAP: Passenger Count through QR Code Scanning

Description

This mobile app allows registered smartphones to collect data on the number of passengers through multiple modes. The app has four modes: QR Code, Face Detector, Manual Counter, and Kiosk.

Requirements

1. Free Expo account
2. Laravel for backend development
3. React Native and Expo for mobile app development
4. Smartphones with camera

The Laravel backend for the Ridemap already supports this app. The endpoint for submission of face detection is /api/qrs. The backend accepts an array of faces with the following attributes: qr_code, station_id, and scanned_at. user_id is extracted from a token used to authenticate the device.

Important: Do not forget to update the api configuration in the utils.js file to point to the url of the API of the backend. Specifically, this line:

export const api = axios.create({
   baseURL: 'https://ridemap-php.herokuapp.com/api'
})

Getting Started

Install the necessary tools:

1. Any IDE (Webstorm is preferred but you can also use Atom and VS Code)
2. Install Nodejs
3. Install Expo
4. Install Expo Go in IOS/Android
5. Install Git

Follow the steps below to get set up:

1. Clone this repository

git clone https://github.com/mlab817/ridemap-mobile.git

2. Change directory to ridemap-counter

cd ridemap-mobile

3. Install dependencies

npm install

4. Update the API endpoint in utils.js file

export const api = axios.create({
   baseURL: 'https://ridemap-php.herokuapp.com/api'
})

5. Start expo dev server

expo start

Or

expo r -c

The second command is used when you cannot connect your device to the webserver.

Follow the instructions in the CLI message to connect your simulator or physical device.

Building and Distributing the App

To build the app, install the Expo Application Services.

1. Create a free account in Expo.
2. Install eas-cli to use eas in command prompt and/or terminal.

npm i -g eas-cli

3. From the root directory, run:

eas build

You can control the build configuration by editing the eas.json file. Open this link for more information

Follow the on-screen instructions. You will find the android/ios bundles in your Expo account under Build menu, e.g. https://expo.dev/accounts/{accountName}/projects/ridemap-counter/builds. You may also integrate submission to Play Store and App Store.

Note: Unfortunately, to build IOS applications, you will need to apply and register to Apple Developer Program which costs $99 yearly. Android build is free and can be downloaded for distribution.

Workflow

Device Authentication

To use this app, devices must be registered in the server. To do this, follow the instructions below:

1. Download the app
2. Open the app to connect to the server for the first time
3. The app will show an invalid device message along with the device ID. Use this device ID to register the device in the server.

Note: The device ID does not refer to the actual device ID of the device but rather the device ID of the app tied to the server. This is unique for every app that is installed in the device.

Read more here: Android and iOS

Using the App

1. Open app
2. Device connects to server
3. Server verifies that device is registered
   1. If device is not registered, display invalid device message along with the device ID. Note that the device ID is not the actual device ID but rather the ID of the device tied with the app.
   2. If device is registered, proceed
4. Prompt user to select station where it is located
5. Put the smartphone in the best position to start scanning faces
6. Start scanning

The app will send data for every batch of faces detected.

Table Structure

- users

Stores data of users of the system

Attribute	Type	Description
id	int autoincrement	Primary key of the table
name	varchar	Name of the user
email	varchar	Email of the user
email_verified_at	timestamp	Timestamp when the user validated their email
password	varchar	Hashed password of the user
device_id	varchar	Device ID of the user
created_at	timestamp	Timestamp when the record was saved in the database
updated_at	timestamp	Timestamp when the record was updated in the database

- stations

Stores data of stations in EDSA Busway

Attribute	Type	Description
id	int autoincrement	Primary key of the table
name	varchar	Name of the station
created_at	timestamp	Timestamp when the record was saved in the database
updated_at	timestamp	Timestamp when the record was updated in the database

- passenger_qrs (QR Code)

Stores data on scanned QR codes

Attribute	Type	Description
id	int autoincrement	Primary key of the table
station_id	int	Foreign key referencing stations table
qr_code	varchar	String representation of QR code scanned
scanned_at	timestamp	Timestamp when the QR code was scanned
user_id	int	Foreign key referencing users table
created_at	timestamp	Timestamp when the record was saved in the database
updated_at	timestamp	Timestamp when the record was updated in the database

- faces (Face Detector)

Stores data on scanned QR codes

Attribute	Type	Description
id	int autoincrement	Primary key of the table
face_id	int	Face ID auto-generated by the device when the face was detected
station_id	int	Foreign key referencing stations table
scanned_at	timestamp	Timestamp when the entry was generated
user_id	int	Foreign key referencing users table (user that submitted the data)
created_at	timestamp	Timestamp when the record was saved in the database
updated_at	timestamp	Timestamp when the record was updated in the database

- passenger_counts (Manual Counter)

Stores data on scanned QR codes

Attribute	Type	Description
id	int autoincrement	Primary key of the table
station_id	int	Foreign key referencing stations table
passenger_in	int	Count of passengers entering the vehicle
passenger_out	int	Count of passengers leaving the vehicle
scanned_at	timestamp	Timestamp when the entry was generated
user_id	int	Foreign key referencing users table (user that submitted the data)
created_at	timestamp	Timestamp when the record was saved in the database
updated_at	timestamp	Timestamp when the record was updated in the database

- kiosks (Kiosk)

Stores data on scanned QR codes

Attribute	Type	Description
id	int autoincrement	Primary key of the table
origin_station_id	int	Foreign key referencing stations table (where data was collected)
destination_station_id	varchar	Foreign key referencing stations table (chosen destination)
captured_at	timestamp	Timestamp when the entry was generated
user_id	int	Foreign key referencing users table
created_at	timestamp	Timestamp when the record was saved in the database
updated_at	timestamp	Timestamp when the record was updated in the database

Screenshots

1. Screen 1 - Splash screen
2. Screen 2 - Device ID - use this to register the device
3. Screen 3 - Request permission to use camera
4. Screen 4 - Select station where device is located
5. Screen 5 - Menu - Select which mode to use to collect data
6. Screen 6 - QR Scanner
7. Screen 7 - Face Detector
8. Screen 8 - Manual Counter
9. Screen 9 - Kiosk

For qr scanner and kiosk, data is submitted for every 10 scans or can be triggered manually. Face Detector and Manual Counter are submitted manually.

Limitations!

• One major limitation for scanning QR code are the logistics involved with it. Since it multiplies the number of QR scans by as much as 50x per vehicle (for buses) compared to RAMA, this may take more than the assigned dwell time for vehicles (5s scan per passenger x 50 = 250s or 4 min 10s). Average dwell time ranges from 1 - 1.5 mins. Further, there's a need to assign more inspectors to operate the QR scanners, distribute QR codes, and collect them afterwards.

• For face detector, a major factor here is the quality of the device, the environment, and the targets.

• For manual counter, a time and motion study must be done to ensure how quickly and accurately transport inspectors can manually count the passengers boarding and leaving the vehicle.

• For kiosk, a major factor is ensuring that passengers do not enter their destination station twice. Security is also a major consideration here esp. if the devices will be left behind and set up.

Proper testing and simulation can help address some of these through improved protocols.

Author

This app is developed by Mark Lester Bolotaolo.