SurveillanceBot Mk3

Overview

SurveillanceBot Mk3 is a state-of-the-art robotic surveillance platform that integrates live video streaming, object detection, GPS tracking, and environmental sensing. Built with ESP32 microcontrollers and FreeRTOS, it enables multitasking and efficient resource management, making it a versatile tool for surveillance and monitoring applications.

---

Evolution of the Project

Mk1: Initial Prototype

The first iteration of SurveillanceBot featured:

• GPS Tracking: Integrated NEO-6M for real-time location tracking.
• Environmental Sensing: Equipped with MQ-2 gas and HX710B pressure sensors.
• Live Video Streaming: ESP32-CAM for real-time camera feeds.
• Remote Control: Controlled via a web-based interface.

While comprehensive, Mk1's performance was limited by the computational load of running all features simultaneously.

*Click to watch the Mk1 demo.*

---

Mk2: Streamlined Performance

To improve performance, Mk2:

• Removed GPS Tracking and Environmental Sensing to reduce computational strain.
• Focused solely on Live Video Streaming and Object Detection using the MobileNet SSD model.
• Optimized the hardware and software for smoother operation.

This iteration achieved significant improvements in stability and responsiveness.

*Click to watch the Mk2 demo.*

---

Mk3: Comprehensive Surveillance

Mk3 reintroduced features from Mk1, addressing earlier computational limitations by incorporating FreeRTOS for multitasking. Key enhancements include:

• Reintegration of GPS Tracking and Environmental Sensing.
• FreeRTOS Multitasking: Efficient task scheduling for live video, object detection, GPS, and sensors.
• Improved Web Interface: Unified control dashboard for all functionalities.

This iteration represents the culmination of the project's evolution, balancing performance and functionality.

---

Features

1. Live Video Streaming and Object Detection:

   • Real-time streaming using ESP32-CAM.
   • Object detection powered by MobileNet SSD and OpenCV.

2. GPS Tracking:

   • Tracks real-time geographic location with the NEO-6M GPS module.
   • Displays location on a web-based dashboard.

3. Environmental Sensing:

   • Detects gas concentrations with the MQ-2 sensor.
   • Measures atmospheric pressure using the HX710B sensor.

4. Multitasking:

   • Manages video, GPS, and sensor monitoring simultaneously using FreeRTOS.

5. Remote Control:

   • Navigate the robot via a web-based interface.
   • Adjust mounted components using a servo motor.

---

Hardware Components

Component	Purpose
ESP32-CAM	Captures live video and performs object detection.
ESP32 DEVKIT	Handles GPS tracking and sensor data monitoring.
NEO-6M GPS Module	Tracks real-time geographic location.
MQ-2 Gas Sensor	Detects gas concentrations in the environment.
HX710B Pressure Sensor	Measures atmospheric pressure.
L298N Motor Driver	Controls the movement of dual-shaft BO motors.
Dual Shaft BO Motors	Provides mobility for the bot.
Servo Motor	Adjusts mounted components remotely.
Battery Pack	Powers the robot's components.

---

Software Requirements

1. Arduino IDE

• Install libraries:
  • esp_camera
  • WiFi
  • TinyGPSPlus
  • HX710B
  • Servo

2. Python

• Install dependencies:

  pip install opencv-python imutils requests flask

3. FreeRTOS

• Utilized for efficient multitasking on ESP32.

---

Circuit Diagrams

Car Control Circuit

---

Tracker Circuit (GPS and Sensor Module)

---

Installation

1. Clone the Repository

git clone https://github.com/Jjateen/SurveillanceBotMk3.git

2. Configure Arduino Sketches

1. Open arduinoSketch/arduinoSketch.ino in Arduino IDE.
2. Set your Wi-Fi credentials in the sketch.
3. Upload the sketch to the ESP32-CAM and ESP32 DEVKIT boards.

3. Run the Python Server

Launch the Flask web server:

python app.py

---

Usage

1. Power On the Bot

• Connect ESP32 modules to the battery pack.

2. Access Web Interfaces

• Use the IP addresses displayed in the serial monitor to access:
  • Live video streaming and robot control.
  • GPS location and environmental sensor data.

3. Control the Bot

• Use the web interface to:
  • Move the robot forward, backward, left, and right.
  • Monitor real-time video, GPS location, and sensor readings.

---

Detailed Working

Multitasking with FreeRTOS

• Task 1: Web server for live video streaming and object detection.
• Task 2: Web server for GPS tracking and environmental monitoring.

Object Detection

• Utilizes MobileNet SSD for identifying objects in the live video stream.
• Processes frames directly on the ESP32-CAM.

GPS and Sensor Monitoring

• Displays real-time GPS coordinates and sensor data on a Flask dashboard.

---

Output and Interface Views

Controller Interface

---

Sat View Interface

---

Flowchart

---

License

This project is licensed under the Apache 2.0 License.

---

References

• SurveillanceBot Mk1 Repository
• SurveillanceBot Mk2 Repository
• FreeRTOS Documentation
• TinyGPSPlus Library

---