3rd year project

Wild Life Tracker

Project Repositry

Introduction

waves

Problem Overview

The traditional method of wildlife researching in Sri Lanka, and many other developing as well as developed countries in the world is to stay in wild areas for a long time, observe animals randomly, and identify areas that animals are most active. This method is ineffective because some animals hide when they sense humans. They don't behave naturally around humans. Thus, it consumes a lot of time to observe their natural behavior. Researchers have to stay in the wild throughout this long period. So, this is a formidable process. Sometimes their life is in danger because there is a possibility to face an attack by wild animals while they are in the jungle. An immense amount of time, money, and human resources are spent in vain for traditional tracking of wild animals for researchers.

waves

Solution

Our solution to this problem is a,

  • Remotely controlable
  • Real time
  • Efficent and effective

tracking system.

This system consists of three major components, A hardware unit, a cloud server, and a web App. The hardware unit consists of camera traps, PIR sensors, and a location tracker. One unit of these three components makes a station. Researchers need to set up this station in the researching area. One researcher can have more than one station established in different places according to their choice. The sensors in a station can detect animals when they are in the sensing range and it will trigger the camera. A real-time photo is captured at that moment and they are stored in the station itself and also sent to the cloud server. Then the researcher can analyze these observed data through the web app.

  • First One

    Remotely controlable

    Researcher can monitor the station through the web app after it is successfully established in researching area.

  • second one

    High Performance Camera Module

    A powerful camera unit to capture perfect photos in any lighting condition. Also provides video recording facilities, recorded videos are stored in the own memory of station.

  • third one

    Tracking System

    Pinpoint the exact location of the station. The most active station can be observed by the Web App.

  • fourth one

    Solar Power

    The stations are powerd by battaries. A solar cell system is used to recharge them.

  • fifth one

    Durable

    Can be set up in any environment. Made with highly durable and waterproof materials so that animals and unforgiving weather can't damage it.

  • sixth one

    Self Storage In Stations

    Data obtained by the system is stored in itself when there is no connection with the database.

NEXT GENERATION OF WILDLIFE RESEARCHING!

**********************

Project Design

Web Application

Organization of the Web App

First

 Here are the two registration pages in our web application.
   01. The user registration Page
   02. The admin registration Page
 Form validations and feedbacks are used to give a better user experience. Clients have to fill all the fields correctly in the forms to enable the submit button. The button provided in left top corner directs user to the home page.

 Here are the two login pages in our web application.
   01. The user login Page
   02. The admin admin Page
   02. The error message if the credentials are not valid.
 Form validations and feedbacks are used to give a better user experience. The user have to provide the correct credential to login to the system.
 If the credentials are not matching it will display the error message. With error message, links back to the login page or password recovery are provided.
 Also in the login pages we have provided a link to password recovery. Users can reset their password using this link.

These are the pages in user dashboard.
There is a home tab, photos tab, location tab and a contact tab.
In photos tab user can see the photos taken by devices connected to his profile. Also there is a option to connect new devices with the profile.
The map provided in location tab pinpoints all the location of devices connected to users profile.


Here are the pages in admins dashboard. The user tab will show all the user requests to the system.
Admin can see details and the letter uploaded by each user by clicking on a request in the list. Then they can submit or reject the request. If they are rejecting the request, they have to provide the reason for the rejection in given text box.

Second

 This is the home page of our web application. Clients can login to there accounts by click on the log in button. The get started button directs clients to user registration form.
 The button provided in the right bottom corner provides a link to contact us via emails.

 There are two pages for password recovery process. The first one is for reqest a password recovery. Here, user have to enter the email of his account and submit.
 If the entered email is valid, it will show up the success message. Else, it will showup the error message.

Database Schema

database
mongo

 We use MongoDB as our database managment system. Because of its scalability, its powerful querying capabilities and the fexibility it provides for modele and manipulate data structures.

Cloud Deployment

  We have used AWS EC2 instance to deploy our web application. NGINX is used as the load balancer And revers proxy of our frontend because of its capability of handling high volume of connections. This helps our web application to handle number of clients at once. Since we have used NodeJS to implement the backend of our web aplication we have used PM2 process manager to keep our nodeJS application alive at all the times.
  Because of the limitations of AWS Educate accounts we were not able to use R3 services to get a domain name. Therefore we had to use seld signed certificate to serve HTTPS.
  We have used MongoDb Atlas as our database managment system. Therefore we have connected the database to server running in EC2 instance.
 Devices are sending data to AWS IoT core using MQTT. These data are stored in a S# bucket. Then the server running in EC2 instance is accessing these data. We were not able to implement this part yet because the AWS educate account dont have permissions to use IAM services. Therefore, at this point we have created the file structure in the server itself.

Hardware

Organization of the Hardware System

hardware diagram

Hardware Components

First

 This unit handles the sensing and taking snapshots of wild animals accordingly. It is consist of ESP32 Cam Board, PIR motion sensor(HC-SR501), Flash LED and a LDR. PIR sesor is capable of sensing motions within 7 meters and 120° range. This sensor will trigger the camera to take photo and then a 30 second video is captured.
 ESP32 cam board is consist of a powerfull image processing unit that allows to take very clear photos and a SD card slot. The PIR sensor is directly connected to the ESP32 cam board. The captured photos are stored in the SD card before send to the cloud database. The captured photo is encoded with Base64 and sent to sim800L module in the connection unit using serial transfer method.

Implementation

Schematic Diagram

hardware schematic

Implementation Details

Libraries Used in Firmware implementation
  • esp_camera.h
  • Arduino.h
  • FS.h
  • SD_MMC.h
  • soc/soc.h
  • soc/rtc_cntl_reg.h
  • driver/rtc_io.h
  • SerialTransfer.h
  • EEPROM.h
  • SoftwareSerial.h
  • TinyGPS++.h
Special Design Decisions
  • SIM800L module is connected to ESP32-CAM board throgh serial port of the ESP32-CAM board. The captured image is sent to SIM800L module as a Base64 encoded byte stream. Also AT commands to control the SIM800L are sent sirially throgh the serial port.
  • The NEO-6M GPS module continously publish GPS data throgh its serial port. So, we needed another serial port in ESP32-CAM module. But there is only one serial port in ESP32-CAM board. Therefore, we have used SoftwareSerial library to configure two GPIO pins as a serial port. We have made GPIO2 and GPIO14 pins as TX,RX for GPS module.
  • The ESP32-CAM board is programmed to go to the deep sleep mode after capturing and sending a image to reduce the power consumption of the device. An interrupt generated by PIR sensor wakes the ESP32-CAM board up.
  • All the photos captured by the camera are written to SD card to make sure there is no data losses.
  • GPS data is updated every 2 hours.
Final Product

Security Features

hardware diagram

Our system handles really sensitive information about wildlife. Therefore, the CIA triad (Confidentiality, Availability, Integrity) of the application is maintained in the best way possible.

  • JsonWebTokens (JWT) are used for the authorization of users. These tokens are stateless, portable, high performing and decentralized which improves the security and scalability of the web application as well.
  • HTTPS protocol is used for secure communication.
  • A special system of user registration is introduced to ensure the users of our web app are researchers.

Testing Plan

Unit Testing

  • Test functionality of each page of the web app
  • Test login, registration and form validations
  • Test sensors and hardware components

Integration Testing

  • Test front end and back end intergration.
  • Test database and back end intergration.
  • Test web app and hardware device intergration.

End to End Testing

  •  Test overall functionality of the web app and hardware device.
  • Check user registraton, registration aproval.
  • Check data flow from devices to user profiles.
  • Check data flow from devices to cloud.

Testing Tools

Front End Testing

selenium

We use selenium automated testing framework to test the front end of our web app. Main reasons to pick selenium is it supports different browsers and platforms. Also, it supports testing scripts written in multiple languages.

We are planing to use selenium to test, form submissions, form validations ,response handling functionalities in our web app.

Back End Testing

postman

We use insomnia API testing tool to test the back end of the web application. insomnia allows us to send HTTP requests using a graphical user interface. insomnia acts as the client and it is possible to analyze server responses to HTTP requests.

We are using insomnia to test, routings implemented in backend, authentication and authorization, error handlings and sending responses in the API.

Testing Procedure and Results

Unit Testing


Front end testing


  As an unit test we tested the functionality of the front end of our web application. This is very important because the front end is used by our clients. So, it must provide all the expected outcomes correctly to give a better user experiance. Therefore, we needed to make sure that the front end of our web application works fine.
  We have used the automate testing tool selenium to test front end. The first test we have done is the navigation test. We wrote a script to click the all possible links inour web application and the we have checked whether it navigate to expected page. Form validations are another important aspect in UI. So, we wanted to test whether our app provides correcet feed backs when the client fill a form. For this also we wrote a selenium script. In this script, we tested all of our forms with invalid data to make sure they are giving correct feed back. Also, we tested the froms with valid data to make sure that the forms are not giving any invalid feedback wheth client inputs valid data.
  The scripts are attached here. The testing results are listed below.


Back end testing


  Prior to the intergration of frontend and backend, we wanted to make sure that our rest API returns correct responses to clients requestes. We use 'Insomnia' to test the API. We have implemented lot of routes in our API to do various tasks and to handle various requests from the frontend of our web application. It is very important to make sure that the API is providing correct responses. Because the security of the web application is depends on these responses. Using insomnia we have tested each route in our API for valid and invalid data.
  We have tested whether the API responds with correct status codes, data and error messages for different data sets. Also as an security test, we have passed wrong credentials to the API using insomnia and checked whether it rejects those requests. We have used JWT tokens to authorize clients. It is very important to make sure tha API is rejecting invalid tokens. Therefore we have tested whether the API is rejecting invalid tokens by sending invalid token using insomnia. All of these tests are conducted manually. The below figure contains all the cases we have tested using insomnia.

insomnia_tests

Integration Testing


  To test the integration of the backend, frontend and the database, we used selenium. We have tested the response from the server to logins, user registrations, password recoveries and invalid inputs to make sure that the data flow from front end to back end and from back end to data base is fine. We have wrote scripts to enter valid and invalid credentials in login pages and check the response of the server.
  If these compnent intergrations are fine, the logins for valid credentials must be successfull. We have tested whether the request with correct credentials directs the client to the dashboards in the scripts. We had to check the database manually to make sure all the data entered by the script is recorded in database correctly.
  Also, we have tested the same script for three popular browsers. Google Chrome, Microsoft edge and Operamini to make sure our web application is ccompatible with different platfoorms.
  The testing scripts are attached here.


Testing Results


Results of tests by Selenium


Test run No 1: Chrome Browser
Passed: 6 Faild: 1

Test run No 2: Chrome Browser
Passed: 7 Faild: 0

Test run No 1: Microsoft Edge Browser
Passed: 7 Faild: 0

Test run No 2: Microsoft Edge Browser
Passed: 6 Faild: 1

Test run No 1: Operamini Browser
Passed: 6 Faild: 1

Test run No 2: Operamini Browser
Passed: 7 Faild: 0
Test Case Chrome Edge Opera
Navigations to all the pages passed passed passed:1 Faild:1
Form validations for incorrect inputs passed:1 Faild:1 passed:1 Faild:1 passed
Form validations for correct inputs passed passed passed
User login and dashboard function passed passed passed
User registration test for valid and invalid data passed passed passed
Admin login and dashboard function test passed passed passed
Password recovery test for valid and invalid emails. passed passed passed

Results of tests by Insomnia

Test Case Returned Status Code Expected Status Code Observation Faild / Passed
User and Admin login for correct credentials 200 200 Backend responds with status code 200 and send a token to client passed
User and Admin login for incorrect credentials 401 401 Responds with status code 401 and rejects the request. Just recived credentials invalid message. passed
User and Admin Registration for emails that are already registered in system 409 409 Rejects registration for same emails. passed
Password recovery for emails that are not registered in system 401 401 Rejects password recovery for invalid emails. passed
Password recovery for emails that are registered in system 201 201 Recived message indicating that the password recovery email is sent to entered email. passed
Try to reset password for second time using a link used before. 401 401 Recived message indicating that the password recovery link is invalid or expiered. passed
Try to reset password of a accont using a token sent to another email 401 401 Recived message indicating that the password recovery link is invalid or expiered. passed
Try to register as an admin using a link that already used to register. 401 401 Recived message indicating that the link is invalid or expiered. passed
Try to log in to the admin dashboard using a token recived for user login 401 401 Recived message indicating that the client is not an admin. Redirection rejected. passed
Confirm or Reject request with admin token. 200 200 With admin token. Possible to confirm or reject client request. passed
Confirm or Reject client request with user token. 401 401 With user tokens. Cannot confirm or reject requests. Recieved message unauthorized. passed

Project Budget

(Budget is made for 1 Hardware unit only)

Hardware Component Amount Price(Rs) Place
Infrared PIR motion Sensor 1 445.00 duino.lk
ESP32 Cam Board 1 2590.00 duino.lk
FTDI connector 2 1365.00 duino.lk
SIM800L module 1 - Purchased from department
NEO-6M GPS module 1 1950.00 duino.lk
Solar Panel (>6V) 165mA 1 - Purchased from department
3.7V 18650mAh batteris 2 1780.00 duino.lk
Battery Holder 1 155.00 duino.lk
Enclosure Box - 1290.00 duino.lk
Other Expenditures - 325.00
TOTAL 9900.00

Project Timeline

time line

Supervisors

Dr. Isuru Nawinne
Dr. Isuru Nawinne
Dr. Mahanama Wikramasinghe
Dr. Mahanama Wikramasinghe

Group Members

Group 9

Esara Sithumal
E/17/176
Kumara W.M.E.S.K

Hansa alahakoon
E/17/006
Alahakoon A.M.H.H

Lakshitha Srimal
E/17/338
SRIMAL R.M.L.C