🚁 Multi-Agent Drone Control System 🚁

📋 Overview

This repository contains MATLAB implementations for multi-agent drone swarm control with obstacle avoidance capabilities and dynamic formation control. The codebase implements various control strategies for drone swarms operating in environments with static and moving obstacles.

🌟 Features

• 🔄 Dynamic formation control using Voronoi tessellation
• 🧭 3D path planning with obstacle avoidance
• 🚧 Collision avoidance between agents
• 🏙️ Building/wall constraint handling
• 📊 Visualization tools for drone trajectories and Voronoi diagrams
• 🛡️ Multiple obstacle detection and avoidance strategies

🗂️ Repository Structure

The repository is organized into three case studies, each exploring different aspects of multi-agent control:

CASE STUDY 1

• Basic drone formation control without obstacle avoidance
• Voronoi-based centroid calculation for optimal coverage
• Building/wall constraint handling

CASE STUDY 2

• Formation control with static and moving obstacles
• Enhanced collision avoidance between agents
• Sensor field-of-view (FOV) based obstacle detection

CASE STUDY 3

• Full 3D control for drone swarms
• Advanced obstacle avoidance using potential fields
• Spherical obstacle representation and avoidance

🛠️ Key Functions

• Centroid_Calculator.m - Computes optimal agent positions using Voronoi tessellation
• dynamics_*.m - Main simulation files for different scenarios
• TransMat.m - Coordinate transformation between reference frames
• Dyn.m - Dynamic model of drone agents
• distance.m - Distance calculation between agents/obstacles

🔬 Mathematical Background

The implementation is based on several key algorithms:

• Lloyd's algorithm for Voronoi-based coverage control
• Artificial potential fields for obstacle avoidance
• Consensus-based flocking control
• Proportional-derivative (PD) controllers for formation maintenance

🚀 Getting Started

Prerequisites

• MATLAB R2020b or newer
• MATLAB Robotics Toolbox (recommended)

Running a Simulation

1. Clone this repository
2. Open MATLAB and navigate to the repository folder
3. Choose a case study folder and run the corresponding dynamics_*.m file
4. Visualize the results using the generated plots

Example:

cd 'CASE STUDY 2'
dynamics_multiobs

📈 Visualization

The code includes visualization tools to display:

• 3D trajectories of all agents
• Voronoi diagrams of agent distribution
• Obstacle positions and avoidance maneuvers
• Relative distances between agents (for collision avoidance verification)

This project is licensed under the MIT License - see the LICENSE file for details.