🤖 3-DOF Robot Arm Path Planning (RRT & RRT*)

A Python-based simulation and benchmarking tool for 3-Degree-of-Freedom (3-DOF) robot arm path planning in a 2D Configuration Space (C-Space). This project implements and compares the standard Rapidly-exploring Random Tree (RRT) and its asymptotically optimal variant, RRT*.

✨ Features

• Configuration Space Planning: Solves the inverse kinematics ambiguity by mapping 2D workspace obstacles into a 3D joint-angle C-Space.
• Interactive GUI: Built with Pygame, offering a real-time visualization of the tree growth and the final smoothed robot arm movement.
• Headless Benchmarking: Includes a Monte Carlo testing suite to evaluate algorithm performance across randomized environments without GUI overhead.
• MVC Architecture: Fully decoupled codebase separating the physical environment, planning algorithms, and UI rendering.

📁 Project Structure

The project follows a modular design for maintainability:

• config.py: Centralized configuration for magic numbers, robot kinematics, and hyper-parameters.
• environment.py: Pure mathematical logic for Forward Kinematics and collision detection (distance from segments to circles).
• planner.py: The core algorithmic implementations of RRT and RRT*, including Rewire logic and path interpolation.
• gui.py: The presentation layer using Pygame for rendering trees, paths, and robot entities.
• main.py: The main entry point supporting both GUI Demonstration and Headless Benchmarking modes.

🚀 Installation & Usage

Dependencies

Ensure you have Python 3.8+ installed. Install the required libraries:

pip install numpy pygame pandas

Running the ProjectOpen main.py and modify the execution mode at the bottom of the script:Mode A: Interactive GUI DemoUncomment the GUI lines to visualize the planning process:

if __name__ == "__main__":
    env = RobotArmEnv(LINK_LENGTHS, CENTER, DEFAULT_OBSTACLE)
    app = RobotArmGUI(env)
    app.run()

Mode B: Headless BenchmarkUncomment the benchmark line to run 50 randomized trials and generate statistical data:

if __name__ == "__main__":
    env = RobotArmEnv(LINK_LENGTHS, CENTER, DEFAULT_OBSTACLE)
    run_randomized_benchmark(env, trials=50)