Skip to content

globaledgesoft/rb5-human-tracking

Repository files navigation

Table of Contents

About The Project

This project is intended to build and deploy a human tracking application onto Qualcomm Robotics development Kit (RB5) that identifies the person by face recognition.

The application consists of 2 parts: A webserver and an MQTT client. Web Server is responsible for storing the faces in RB5. It waits for POST Request from client.

API Info: Request Type: POST Endpoint name: /add_face Image can be included as multipart form data with field name as “image” Eg. (HTML Form)

MQTT Client subscribes to topic “track” and expects the name of the person as payload. You can test it using Mosquitto clients by entering the following command:

mosquitto_pub -m "<name of the person to track>" -t "track"

Once a name is published to “track” topic, RB5 will initiate the tracking process for a particular person. Once the person is identified, RB5 will publish the tracked information to the topic “tracked” with payload containing following information: { “name”: , “score” : , “location”: }

Prerequisites

  1. A Linux workstation with Ubuntu 18.04.

  2. Install Android Platform tools (ADB, Fastboot)

  3. Download and install the SDK Manager

  4. Flash the RB5 firmware image on to the board

  5. Setup the Network

  6. Setup the Pybind11.

  7. Setup opencv for python on RB5

  8. Turtlebot burger is assembled, operational and is connected to RB3

Getting Started

To get a local copy up and running follow these steps.

  1. Clone the project repository from the github to RB5.
git clone 

Steps to install additional python libraries

1.Install pybind11

python3.5 -m pip install pybind11
  1. Install opencv for python
python3.5 -m pip install opencv-python
  1. Install Flask
python3.4 -m pip install Flask
  1. Setup mosquitto Broker
sudo apt-get update
sudo apt-get install mosquitto

Test the broker by installing mosquitto client

sudo apt-get install mosquitto-clients

Now subscribe to test topic publish to that topic and check whether subscriber got the message

mosquitto_sub -t "test"
mosquitto_pub -m "message from publisher" -t "test"

Usage

Follow the below steps for initializing the application:

  1. Start the web server
python3.5 server.py
  1. Start the MQTT client for tracking:
python3.5 main.py

Contributors

About

To build and deploy a human tracking application on RB5, that identifies the person by face recognition

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages