Skip to content

README5.md

Latest commit

 

History

History
302 lines (171 loc) · 11.9 KB

README5.md

File metadata and controls

302 lines (171 loc) · 11.9 KB

Coyote Team 1: IoT & Network

LoRaWAN implementation for acoustic localization of coyotes

Problem Statement

In United States, cases of human injuries, loss of pets and livestocks caused by wild coyotes are reported every year. Since wild coyotes usually inhabit in rural area, pets and livestocks in those places are more vulnerable to coyote attacks. However, there are few specialized solution or devices created to solve this problem.

Novelty

LoRaWAN

Similar studies whose purpose is to detect animals use WiFi. However to cover large areas such as farmland, it's range of communication is too short. (45 meters at 2.4GHz, 15 meters at 5GHz) Also devices for its communication have high cost, making it totally unaffordable to use in any agricultural solution. This is why LoRaWAN is implemented in this project. It has wider communication range, especially in rural area. (5 kilometers in urban area, 15 kilometers in rural area) And the equipment cost is lower than WiFi.

Project Overview

2022-10-15_architecture

System Flowchart

2022-10-15_architecture

Environment Settings

Sensors

  • Heltec ESP32 WiFi LoRa 32 (V2)

  • Raspberry Pi v3

  • USB microphone AK5371

End Node Settings

2022-10-15_architecture

Microphone - Raspberry Pi - ESP 32

code/rec_voice_ver1.py should be run inside the Raspberry Pi.

  • dependencies

      #install sounddevice ver 0.4.1
  python3 -m pip install sounddevice
  
  #install pyserial ver 3.4
  python -m pip install pyserial

The microphone(acoustic sensor) needs port setting due to the usb port changing issue. It happens when the Raspberry Pi reboots, so it needs to be fixed to a custom port ttyUSB_DEV1. Following is the setting process of the Raspberry Pi.

  • USB port setting

      check ttyUSB port using
      $ ls /dev
  check idProduct value using
      $ udevadm info --name=/dev/"checked tty usb port" --attribute-walk
  check rules file
      $ sudo nano /etc/udev/rules.d/**
  write this in the rules file
      SUBSYSTEM=="tty", ATTRS{idProduct}=="idProduct value checked before", SYMLINK+="ttyUSB_DEV1"
  run
      $ sudo udevadm trigger
  check the new fixed port
      $ ls -l /dev/ttyUSB*

ESP 32 - Arduino IDE Settings

The code is also dockerized, check code/end_node/esp32_send_dockerized to compile the code in docker. .ino file is in volumes/Arduino/esp32_send. You have to change the APPKEY, DEVEUI, APPEUI for each sensor.

To compile the code, run the following command in /esp32_send_dockerized

    docker build -t esp32_send .

To run the docker image

    docker run -it --rm -v $PWD/volumes/Arduino:/Arduino esp32_send

  • To stop the container, type in

      exit

  • inside the container

To compile the code, type in

    arduino-cli compile --fqbn esp32:esp32:heltec_wifi_lora_32_V2 --verbose esp32_send.ino

Inside the container, and the code in /volume/Arduino/esp32_send will be compiled

To upload the code, the device must be connected to the computer, and device should be connected with docker

To connect the device with docker, for ubuntu, type in

    docker run -it --rm -v $PWD/volumes/Arduino:/Arduino --device=/dev/ttyUSB0 esp32_send

  • Then, inside the container, type in

      arduino-cli upload -p /dev/ttyUSB0 --fqbn esp32:esp32:heltec_wifi_lora_32_V2 --verbose esp32_send.ino

  • For the other OS, please refer to this page

Gateway

RAK7249 WisGate Edge Max from RAKwireless

Network

The Things Stack Community Edition v3.22.1

  • confidentials

      console address: https://nam1.cloud.thethings.network/console/
  
  id: lim406@purdue.edu
  
  pw: featherC9130*
  
  (expired by 2022.12.22)

Localization

This code is working on the premise of the fact that all of timestamp data from three sensors are received and the time difference between them are less than certain threshold described in the code. Otherwise, the code will loop and nothing will show up in the terminal until it gets the right input from the things stack.

  • dependencies

    paho-mqtt

      pip install paho-mqtt

    sympy

      pip install sympy

  • mqtt subscribe settings

    documentation link: https://www.thethingsindustries.com/docs/integrations/mqtt/mqtt-clients/eclipse-paho/

    change mqtt subscribe settings to connect to your own tts server

      sub = subscribe.simple(topics=['#'], keepalive=10 ,hostname="your tts server", port=1883, auth={'username':"your application name",'password':"your application's api key"}, msg_count=3)

  • Environment setting

2022-10-15_architecture

2r is the distance between sensors. and sensors are located in the form of an equilateral triangle. time_obj is the timestamp that indicates the detection time. td0 is the difference of 0, 1 and td1 is between 1, 2 and td2 is between 2, 0. v is the speed of sound. the area represents the section sound source is located.

The input you have to change is 2r, (m, n), which are the distance between sensors, and the location coordinates of the center of the triangle. Set those values of your own experiment environment. The default values are 304.8 cm and the location of KSW-square at purdue university.

You need to measure the cardinal direction and put the difference to the value theta. The default value is 8 degrees.

Node.js Server

install docker desktop, at code/server/coyote-main, run

    docker build . -t your username/coyote-main
    docker run -p 8081:8081 -p 3333:3333 -d your username/coyote-main

server will run on your localhost.

http connection (port 8081)

    POST http://localhost:8081/api/sensors/getSound1Coord : get the coordinate information of sound 1 sensor

    POST http://localhost:8081/api/sensors/getSound2Coord : get the coordinate information of sound 2 sensor

    POST http://localhost:8081/api/sensors/getSound3Coord : get the coordinate information of sound 3 sensor

    POST http://localhost:8081/api/coyotes/getInitialCoyotes : get the coordinate information of last 5 detected coyotes

websocket connection (port 3333)

    ws://127.0.0.1:3333

Unity

version

  • 2021.3.9f1

Imported files and assets

Configuration before executing

  1. Make sure to insert the Unity Scene File before running

  2. Unity Editor → File → Build setting → Putting a Assets/Scenes/LoadingScene, Assets/Scenes/Coyote inside ‘Scenes in Build’

  3. Unity Editor → Edit → Project settings → Player → Android → OtherSettings → Identification → Set ‘Minimum API Level’ to Android 7.0 ‘Nougat’ (API Level 24), Set ‘Target API Level’ to Automatic(highest installed)

  4. Unity Editor → File → Build setting → Platform →Set to ‘Android’

  5. If you downloaded MapBox SDK again, change the SDK script DeviceLocationProvider.cs to code/unity/DeviceLocationProvider.cs

  6. How to set up a Mobile Phone Connection Environment for Project Run

    1. Install the Unity Remote5 app on your phone device
2. Go to your phone developer options → Set USB Debugging Permissions to Allow
3. Unity Editor → Edit → Project settings → Editor → Unity Remote → Set ‘Device Settings’ to ‘Any Android Device’
4. Run the Unity Remote5 app on your phone → Connect your phone to your computer (data cable) → Project Run

  7. Android only (version: Android 7.0 'Nougat', API Level 24 or higher)

Error debugging

  • "namespace name does not exist" during MapBox SDK import": an error related to Mapbox AR, resolved after deleting all AR-related files in Mapbox.
  • "warning: the option setting 'android.enabler8=false' is deprecated. it will be removed in version 5.0 of the android gradle plugin. you will no longer be able to disable r8", "starting a gradle daemon, 1 incompatible daemon could not be reused, use --status for details": an error due to android gradle file, resolved after changing version from com.android.tools.build:gradle:3.6.0 to 3.4.0 inside the 'baseProjectTemplate.gradle' file.
  • "manifest merger failed 22 cannot be smaller than version 24 declared in library error": an error related to version error of minSDK and targetSDK in the AndroidManifest.xml in the Unity library file.
  • "Import error: the name 'heading' does not exist in the current context": an error in DeviceLocationProvider.cs script file due to encoding issue. Change encoding method from ISO-8859 to UTF-8 or ASCII.

Group Members

  • Jaehui Boo

    Computer Engineering

    Dankook University

    Yongin, Republic of Korea

    yeppyboo@gmail.com

  • Hyemin Lim

    Dept. of Computer Science and Engineering

    Chung-Ang University

    Seoul, Republic of Korea

    lhye9130@gmail.com

  • Hyeongjun Kim

    Computer Engineering

    Daegu Catholic University

    Gyeongsan-si, Republic of Korea

    kim27978965@gmail.com

  • Nayoun Kim

    Information Technology Convergence

    Woosong University

    Daejeon, Republic of Korea

    nayounkim797@gmail.com

  • Justin Anderson

    Computer and Information Technology

    Purdue University

    West Lafayette, Indiana

    ande1013@purdue.edu

  • WeiChieh Chin

    Computer and Information Technology

    Purdue University

    West Lafayette, Indiana

    victochinr900630@gmail.com