🚗 Two-Camera Autonomous Parking System

An autonomous parking system that uses two fisheye cameras and YOLO-based perception within a ROS2 + Gazebo simulation environment.

---

📌 Overview

This project implements an autonomous parking pipeline using two wide-angle (fisheye) cameras mounted on a simulated vehicle. The system detects parking spots and surrounding obstacles using YOLO object detection and segmentation models, and executes a parking maneuver autonomously.

Key highlights:

• Dual fisheye camera setup for wide field-of-view coverage
• Bird's Eye View (BEV) transformation for top-down spatial understanding
• YOLOv8-based detection and segmentation (120° camera models)
• ROS2 Humble + Gazebo simulation environment
• Pre-trained .pt model weights included

---

🛠️ Tech Stack

Category	Tools / Libraries
Middleware	ROS 2 Humble
Simulation	Gazebo
Vision / AI	YOLOv8 (Ultralytics 8.2.69), OpenCV
Language	Python 3, Shell, CMake
Other	pyserial, pynput, transformers, HuggingFace Hub

---

⚙️ Installation

Prerequisites

• Ubuntu 22.04
• ROS 2 Humble
• Python 3.10+

Setup

1. Clone the repository:

   git clone https://github.com/yunss01/two_camera_parking.git
   cd two_camera_parking

2. Run the installation script:

   chmod +x install.sh
   ./install.sh

   This will install:

   • Python dependencies: opencv-python, pyserial, ultralytics, pynput, transformers, huggingface_hub
   • ROS2 packages: gazebo-ros-pkgs, xacro, gazebo-ros, gazebo-msgs, gazebo-plugins
   • Gazebo simulation assets

3. Build the ROS2 workspace:

   cd ~/ros2_autonomous_vehicle_simulation
   source /opt/ros/humble/setup.bash
   colcon build --symlink-install
   source ./install/local_setup.bash

---

🚀 Usage

Launch the simulation

ros2 launch simulation_pkg mission_sim.launch.py

---

🤖 Pre-trained Models

File	Description
best_120cam_detection_1.pt	Object detection model for 120° fisheye camera
best_120cam_segmentation.pt	Segmentation model for 120° fisheye camera

---