🤖 PSO-RobotPathFinder

📖 Overview

PSO-RobotPathFinder is an intelligent navigation framework designed to solve complex robotic path planning challenges using Particle Swarm Optimization (PSO).

This project goes beyond simple A* or Dijkstra implementations by using evolutionary algorithms to generate continuous, smooth, and kinematically feasible cubic spline paths. It effectively balances the trade-off between path shortness and safety margins, ensuring robots can navigate narrow corridors and obstacle-rich environments without collision.

Ideal for autonomous mobile robot (AMR) researchers, students, and developers looking for a Python-based, visualized approach to path optimization.

✨ Key Features

• 🧠 particle Swarm Optimization (PSO): Efficient global search algorithm for finding optimal control points.
• 📐 Cubic Spline Smoothing: Generates kinematically feasible and smooth paths suitable for real-world robot navigation.
• 🖥️ Interactive Streamlit Dashboard: A powerful web-based UI for real-time parameter tuning, environment configuration, and visual debugging.
• 🚧 Dynamic Environment Modeling: Supports adjustable workspace dimensions, start/goal positions, and custom circular obstacles.
• ⚡ Real-time Visualization: Live feedback of the optimization process, showing particle convergence and path evolution.

🛠️ Installation

Prerequisites

Ensure you have Python 3.8+ installed.

1. Clone the Repository

git clone https://github.com/TRIBAK-Mohamed/PSO-RobotPathFinder.git
cd PSO-RobotPathFinder

2. Install Dependencies

It is recommended to use a virtual environment.

pip install -r requirements.txt
# OR manually:
pip install numpy matplotlib scipy streamlit

🚀 Usage

Option A: Interactive Web App (Recommended)

Launch the Streamlit dashboard for a full interactive experience.

streamlit run streamlit.py

Dashboard Capabilities:

• Environment Setup: Adjust width, height, and robot radius.
• Obstacle Editor: Add, remove, or modify obstacles dynamically.
• PSO Tuning: Fine-tune c1, c2, inertia, and population size on the fly.
• Visualization: Watch the solver converge in real-time.

Option B: Desktop Demo

Run the standalone Matplotlib demo for a quick test without the web interface.

python main.py

🧩 Project Architecture

The codebase is modular and organized for scalability:

File/Module	Description
streamlit.py	Main entry point for the interactive web application.
pso.py	Core implementation of the Particle Swarm Optimization algorithm.
main.py	Lightweight desktop demo script.
path_planning/	Core Logic Package
├── solution.py	Handles Cubic Spline generation and path interpolation.
├── environment.py	Defines the workspace bounds and obstacle properties.
├── cost.py	Computes the fitness function (Path Length + Collision Penalty).
└── plots.py	Visualization utilities for Matplotlib.

⚙️ Configuration & Tuning

To achieve the best results, you can tune the PSO hyperparameters:

• Population Size: Higher values improve search thoroughness but increase computation time.
• Inertia Weight (w): Controls the momentum of particles. Higher w favors exploration; lower w favors exploitation.
• Coefficients (c1, c2): Balance the particle's tendency to follow its own best position (c1) vs. the swarm's global best (c2).

🤝 Contributing

Contributions are welcome! Please fork the repository and submit a pull request for any enhancements or bug fixes.

---

Made with ❤️ by TRIBAK Mohamed
📧 mohamedtribak912@gmail.com

without formatting artifacts