Smart Home Multi-Agent System

This repository contains the code for a Smart Multi-Agent System designed to manage a smart home environment using the JADE (Java Agent DEvelopment Framework) and various machine learning models. The system is distributed across three platforms and includes multiple agents representing different IoT devices.

Project Description

The goal of this project is to create a smart home system that leverages multi-agent systems and machine learning to optimize the coordination of services, improve service quality, optimize resource usage, and minimize energy costs.

Agents and Their Roles

1. SmartHomeAgent

   • Responsible for decision-making based on data from other IoT agents.
   • Utilizes machine learning models to make predictions and decisions.

2. AirQualitySensorAgent

   • Represents an air quality sensor.
   • Interacts with SmartHomeAgent.

3. SecurityCameraAgent

   • Represents a security camera.
   • Interacts with SmartHomeAgent and uses a pre-trained convolutional neural network model for facial recognition.

4. HVACAgent

   • Represents a heating/ventilation/air conditioning actuator.
   • Interacts with SmartHomeAgent.

5. LightControlAgent

   • Represents a light control actuator.
   • Interacts with SmartHomeAgent.

6. WateringControlAgent

   • Represents a watering control device.
   • Interacts with SmartHomeAgent.

7. SmartLockAgent

   • Represents a smart lock device.
   • Uses the BARD (Behavior-Aided Reasoning for Chatbots) library for enhanced decision-making.

8. ShutterControlAgent

   • Represents a shutter control device.
   • Interacts with SmartHomeAgent.

9. SmokeSensorAgent

   • Represents a smoke sensor.
   • Interacts with SmartHomeAgent.

Platforms Distribution

• Platform1: SmartHomeAgent, AirQualitySensorAgent, SecurityCameraAgent
• Platform2: HVACAgent, LightControlAgent, WateringControlAgent
• Platform3: SmartLockAgent, ShutterControlAgent, SmokeSensorAgent

Objectives

1. Modeling Agents: Efficiently model agents to represent IoT devices and their capabilities.
2. Inter-Agent Communication: Implement effective communication mechanisms for distributed coordination.
3. Data Collection: Collect data from various sensors for decision-making.
4. Adaptability and Continuous Learning: Enable agents to dynamically adapt to changes and continuously learn from the environment.
5. Machine Learning Algorithms: Utilize suitable machine learning algorithms to predict device behavior and anticipate user needs.

Getting Started

Prerequisites

• Java Development Kit (JDK) 8 or higher
• JADE Framework
• Machine Learning models from Hugging Face
• Apache Maven

Installation

1. Clone the repository:

   git clone https://github.com/AbdoAitrais/ILISI-SMART-HOME-SMA.git

2. Navigate to the project directory:

   cd ILISI-SMART-HOME-SMA

3. Build the project using Maven:

   mvn clean install

Running the Application

1. Start the JADE runtime environment:

   java -cp target/your-jar-file.jar jade.Boot

2. Deploy agents on the specified platforms by running the respective classes.

Usage

• A graphical user interface (GUI) acts as the dashboard for the application and is associated with the SmartHomeAgent.

Contributions

Contributions are welcome. Please fork the repository and create a pull request with your changes.