Smart Waste Collection System and its Digital Twin

This repository implements a Smart Waste Collection System based on a system-of-systems (SoS) architecture for smart city infrastructure management.

Please find the demonstrative video in: Waste Twin demo

Inside the /src folder, the project is divided into two main parts:

• Physical Twin: Code running on physical devices (Raspberry Pi, sensors, actuators, etc.) that controls traffic lights, fleet trucks, and waste bin monitoring.
• Digital Twin: Code developed in Godot that represents the Digital Twin (DT) and visualizes the physical system for digital monitoring and control.

---

Physical SoS Architecture

Below is a figure illustrating the overall SoS. It shows how different components (CS1, CS2, CS3) communicate via MQTT to accomplish the objectives of the SoS and how they are physically connected:

1. CS1 (Traffic Lights):

   • Publishes traffic light status via the topic traffic/lights/status to the MQTT broker located in the CS2.

2. CS2 (Collection Trucks):

   • Contains the MQTT broker.
   • Subscribes to the traffic light status from CS1 and bins data from CS3.
   • Perform actions such as stoping at the red lights and collecting the waste from filled bins

3. CS3 (Waste Bins):

   • Publishes sensor readings to the CS2, which then triggers waste collection of the closest truck.

SoS DT Architecture

Folder Structure

• Physical Twin:
  Contains Python code that runs on Raspberry Pi devices. It comprises three main modules:

  • Traffic Light Control: Manages LED traffic lights and publishes their status via MQTT.
  • Fleet Management: Simulates and controls truck movement using servo motors, MQTT messaging, and GPIO inputs.
  • Bin Monitoring and Disturbance Simulation: Monitors waste bins using ultrasonic and DHT sensors.

• Digital Twin (Godot):
  Contains the Godot project that represent the digital shadows and twin to:

  • Provide real-time visualization of the system’s behavior.
  • Allow for remote monitoring.

---

1. Physical Twin Folder

1. Traffic Light Control Module (src/physicalTwin/lighting/classes.py)

• PTLightingEdge:

  • Entry point for the traffic light system.
  • Initializes the sensor module and starts the traffic cycle.

• PTLightingSensor:

  • Controls the red, yellow, and green LEDs.
  • Publishes the current traffic light status via MQTT.
  • Manages the traffic light cycle (red → green → yellow).

2. Fleet Management Module (src/physicalTwin/fleet/classes.py)

• PTFleetEdge:

  • Main controller for fleet management.
  • Initializes the fleet sensor and starts truck operations.

• PTFleetSensor:

  • Sets up an MQTT client to listen to topics (traffic lights, bin sensors).
  • Monitors GPIO buttons to simulate truck braking.
  • Processes incoming MQTT messages to update truck statuses.

• PTTruck (Base Class):

  • Manages truck movement via a servo motor.
  • Implements acceleration and stop logic.

• PTTruck1 and PTTruck2 (Derived Classes):

  • Specialized truck classes with unique GPIO pin configurations.
  • Run independent control loops using multi-threading.

3. Bin Monitoring and Disturbance Simulation Module (src/physicalTwin/bins/classes.py)

• PTBinsEdge:

  • Entry point for the bin monitoring system.
  • Initializes the bins sensor and starts sensor monitoring.

• PTBinsSensor:

  • Manages the LCD display for showing bin status.
  • Sets up an MQTT client to subscribe to bin disturbance messages.
  • Updates the LCD text and background color based on sensor data.

• PTUltrasonicDHT:

  • Extends the bins sensor to include ultrasonic and DHT sensor readings.
  • Calculates bin fullness and publishes sensor data.
  • Contains a keyboard listener (in a separate thread) to simulate disturbances.

• LCD:

  • Controls the LCD display via I2C.
  • Provides functions for setting text (with/without refresh), changing background color, and creating custom characters.

2. Digital Twin Folder

This folder needs to be imported in GoDot