CARLA Autonomous Driving Simulation with RRT-Based Path Planning and PID Control

Overview

Autonomous driving simulation using the CARLA Simulator🚗. It implements a complete pipeline including:

• Environment Setup: Loading a CARLA world (e.g., Town03), configuring synchronous mode, and spawning vehicles (ego and obstacles).
• Path Planning: Generating collision-free paths using sampling-based algorithms (RRT, RRT*, and Informed RRT*) implemented in RRT.py.
• Trajectory Generation: Smoothing the planned path using motion primitives via cubic polynomial interpolation in motion_primitive.py.
• Vehicle Control: Controlling the ego vehicle using a combination of PID controllers for longitudinal (throttle/brake) and lateral (steering) actions, implemented in controller.py.
• Real-Time Visualization: Displaying a camera view using PyGame and a live-updating trajectory plot using Matplotlib.

File Structure

• main.py:
  Main source code that sets up the simulation, spawns vehicles, attaches sensors, executes path planning, generates trajectories, runs the control loop, and displays real-time visualizations.

• RRT.py:
  Contains the implementation of RRT, RRT*, and Informed RRT* algorithms for path planning.

• motion_primitive.py:
  Implements motion primitive generation using cubic polynomial interpolation for smooth trajectory creation.

• controller.py:
  Implements PID controllers for both longitudinal (speed) and lateral (steering) vehicle control.

  

Requirements

• CARLA Simulator:
  Download and install CARLA from here. Ensure the CARLA server is running before executing the project.

Installation

1. Clone the Repository:

   git clone https://github.com/taherfattahi/carla-motion-planning-rrt-based
   cd carla-motion-planning-rrt-based

2. Install the Required Python Packages:

   pip install numpy matplotlib pygame scipy

Usage

1. Start the CARLA Server:

Run the CARLA simulator (e.g., execute ./CarlaUE4.sh on Linux or run the executable on Windows).

2. Run the Main Script:

python main.py

Contributing

Contributions are welcome. If you find a bug or have a feature request, please open an issue or submit a pull request.

Resources

• CARLA Open-source simulator for autonomous driving research.
• Robotic Path Planning: RRT and RRT*
• The PID Controller & Theory Explained