GitHub - PukyBots/path_planning_gazebo-smart-navigation-challenge: Autonomous robot navigation in Gazebo where the bot reaches a goal while collecting bonus points, avoiding dynamic obstacles, and optimizing time.

PATH PLANNING ROBOT Intelligent Grid Navigation with Vision & avoiding protocol

📌 Overview

An autonomous robot navigation system where a differential-drive robot navigates a 5×5 grid arena, collects bonus points, avoids obstacles, and reaches the goal — implemented in both Gazebo simulation and on a real physical robot.

An overhead laptop webcam handles computer vision (HSV colour detection & obstacle recognition), A* pathfinding plans the optimal route, and commands are sent wirelessly to a Raspberry Pi-powered wheeled robot for real-time execution.

🏆 Scoring System

Bonus sheets are placed randomly in the arena each round.

Sheet Color	Points	Priority
🔴 Red Sheet	10 pts	High
🔵 Blue Sheet	5 pts	Medium
⬛ Obstacle (Cardboard Box)	Penalty / Block	Avoid

🎯 How it works:

An overhead laptop webcam captures a live top-down view of the 5×5 grid arena
OpenCV detects and classifies grid cells using HSV colour detection — red bonus sheets (10 pts), blue bonus sheets (5 pts), and cardboard box obstacles
A* pathfinding computes the optimal route — collecting all bonuses first, then navigating to the goal
The laptop-side vision system sends movement commands over WebSocket to the Raspberry Pi on the robot
The Raspberry Pi translates commands and forwards motor instructions to the Arduino via serial (115200 baud) as 2-byte packets [CMD, SPEED]
The Arduino drives two DC gear motors through the motor driver
A live web dashboard (Flask + WebSocket) streams the warped grid view, score, phase, obstacle count, and event log in real time

🏗️ System Architecture

Laptop (Vision + Planning)
│
├── camera_vision.py       → Overhead webcam, HSV colour detection
├── grid_manager.py        → 5×5 grid state, obstacle/bonus tracking
├── autonomous_bridge.py   → A* pathfinding, decision logic
├── web_dashboard.py       → Flask + WebSocket live dashboard
└── hardware_bridge.py     → Serial communication → RPi/Arduino
         │
         │ Serial (USB) / WebSocket (Wi-Fi)
         ▼
Raspberry Pi (On-Robot)
         │
         ▼
Arduino (Motor Control & I/O)

🧠 Core Features

Feature	Details
🗺️ Path Planning	A* algorithm with bonus-collection-first routing
👁️ Computer Vision	Overhead webcam, HSV-based color detection (OpenCV)
📦 Obstacle Detection	Cardboard box obstacles detected via vision
🎯 Bonus Collection	Red (10 pts) & Blue (5 pts) sheets prioritized in routing
🔌 Serial Communication	RPi ↔ Laptop via `/dev/ttyUSB0` at baud rate `115200`
🌐 Web Dashboard	Flask + WebSocket live grid monitoring
🧩 Modular Design	ROS2 Humble package structure, easy to extend

🗂️ Project Structure

gazebo-smart-navigation-challenge/
├── arduino code/
│   └── arduino.ino                  # Motor control firmware
├── real robot/                       # ROS2 package: path_ws
│   ├── autonomous_bridge.py         # RPi ↔ laptop WebSocket bridge + A* logic
│   ├── camera_vision.py             # Webcam feed & HSV colour detection
│   ├── grid_manager.py              # 5×5 grid state management
│   ├── hardware_bridge.py           # Serial/WebSocket communication
│   ├── stream.py                    # Camera stream handler
│   ├── web_dashboard.py             # Live monitoring dashboard
│   ├── gazebo_tutorials.launch.py
│   ├── package.xml
│   └── setup.py
├── simulation/                       # ROS2 package: gazebo_tutorial
│   ├── __init__.py
│   ├── bonus_grid.world             # Gazebo world file (5×5 arena)
│   ├── bonus_randomizer.py          # Randomises bonus positions each run
│   ├── collector_controller.py      # Simulation robot controller
│   ├── gazebo_tutorial              # ROS2 package entry
│   ├── gazebo_tutorials.launch.py
│   ├── lidar.xacro                  # Robot URDF with RPLidar
│   └── setup.py
└── README.md

🔧 Hardware Stack

Component	Role
Raspberry Pi 4 Model B	Main compute unit on robot
Arduino Uno	Motor control & I/O
Laptop Webcam (overhead)	Computer vision for grid detection
DC Gear Motors × 2	Drive wheels
raspberry type-c adapter	Power supply
2-wheels	to move the robot

🔌 Serial Connection

Parameter	Value
Port	`/dev/ttyUSB0`
Baud Rate	`115200`

💻 Software Stack

Software	Version	Purpose
Ubuntu 22.04 (WSL2)	—	Development OS
ROS2 Humble	Humble	Robot middleware
Gazebo Classic	—	Simulation environment
Python 3	3.x	Main programming language
OpenCV (`opencv-contrib-python`)	4.x	Computer vision, colour detection
PySerial	Latest	Serial communication with Arduino
NumPy	Latest	Array operations
Flask + Flask-SocketIO	Latest	Live web dashboard
Arduino IDE	Latest	Motor controller firmware

📡 Serial Command Protocol

Commands are sent as 2-byte binary packets: [CMD_BYTE, SPEED_BYTE] at 115200 baud.

CMD Byte	Action	SPEED Byte
:--------:	--------	------------
`F`	Move Forward	PWM value (40–220)
`B`	Move Backward	PWM value (40–220)
`L`	Turn Left (burst)	PWM value (40–220)
`R`	Turn Right (burst)	PWM value (40–220)
`S`	Stop	`0`

🏟️ Arena Setup

Property	Value
Grid Size	5×5 cells
Boundaries	Cardboard walls
Obstacles	Cardboard boxes (randomly placed)
Red Bonus Sheets	10 points each
Blue Bonus Sheets	5 points each
Camera Position	Overhead laptop webcam (bird's-eye view)

🚀 Setup & Installation — Simulation

Prerequisites

Ubuntu 22.04 / 24.04 (or WSL2)
ROS2 Humble installed
Gazebo Classic installed

Build & Run

# Navigate to your ROS2 workspace src folder
cd ~/ros2_ws/src

# navigate your workspace root and build
colcon build --packages-select gazebo_tutorial

# Source the workspace
source install/setup.bash

# Launch the simulation
ros2 launch gazebo_tutorial gazebo_tutorials.launch.py

Recording.2026-06-13.132822.mp4

🚀 Setup & Installation — Real Robot

Prerequisites

Raspberry Pi 4 with Raspberry Pi OS
Arduino Uno connected via USB
Overhead USB webcam connected to laptop
Python 3 and pip installed
ROS2 Humble installed

Step 1 — Clone the Repository (on laptop)

git clone https://github.com/PukyBots/gazebo-smart-navigation-challenge.git
cd gazebo-smart-navigation-challenge

Step 2 — Install Python Dependencies

pip3 install opencv-contrib-python numpy flask flask-socketio pyserial

Step 3 — Build ROS2 Workspace

cd ~/path_ws
colcon build
source install/setup.bash

Step 4 — Upload Arduino Code

Open arduino code/arduino.ino in Arduino IDE
Connect Arduino Uno via USB to laptop
Select Tools → Board → Arduino Uno
Select correct port under Tools → Port
Click Upload and wait for Done uploading.
Disconnect from laptop, connect to Raspberry Pi

Step 5 — Find Arduino Port on Raspberry Pi

ls /dev/ttyUSB*

Give permission if needed:

sudo chmod 666 /dev/ttyUSB0

Step 7 — Run the System

On the laptop (vision, A*, dashboard):

ros2 run follow_grid camera_vision

Open the live dashboard in your browser:

http://localhost:5000

On the Raspberry Pi (to launch/start the robot):

ros2 launch follow_grid gazebo_tutorials.launch.py

📸 [PICTURE / VIDEO — Real robot running in arena]

🌐 Web Dashboard

The dashboard (web_dashboard.py) provides:

Feature	Description
📷 Live Feed	Overhead camera feed with warped grid overlay
🗺️ Grid View	Robot position, obstacles, bonus sheets
🛤️ Path Display	Current A* planned path visualisation
📊 Score Tracker	Live bonus points collected
📋 Event Log	Real-time system messages
🔴🔵 Cell Detection	Red/Blue bonus cells and orange obstacle cells

VID_20260612_155937_1.mp4

🛠️ Troubleshooting

Issue	Fix
`Permission denied /dev/ttyUSB0`	Run `sudo chmod 666 /dev/ttyUSB0`
Colours not detected correctly	Adjust HSV values in `camera_vision.py`
Grid misaligned with camera	Re-run perspective warp calibration
Robot not moving	Check baud rate is `115200`, verify USB cable
ROS2 nodes not found	Run `source install/setup.bash` first
Camera window doesn't open	Make sure webcam is plugged in before running

✅ Project Status

Done

Gazebo simulation with 5×5 bonus grid and differential-drive robot
Overhead camera HSV colour detection for grid classification
A* pathfinding with bonus-first routing
Real robot motor control via Arduino serial protocol
WebSocket communication between laptop and Raspberry Pi
Live web dashboard with real-time grid overlay and scoring

📋 To Do

Fine-tune HSV thresholds for varying lighting conditions
Improve motor tick calibration for precise grid movement
Migrate to full ROS2 topic-based architecture for real robot

📋 Project Info

Field	Details
Team	Mohammed Afeez& Moosa Mubasir
Program	TCE Internship — Sahyadri College of Engineering & Management
Mentor	Pulkit Garg, Technical Career Education
Organization	TCE(Technical Career Education)

Navigation Menu

Repository files navigation

📌 Overview

🏆 Scoring System

🎯 How it works:

🏗️ System Architecture

🧠 Core Features

🗂️ Project Structure

🔧 Hardware Stack

🔌 Serial Connection

💻 Software Stack

📡 Serial Command Protocol

🏟️ Arena Setup

🚀 Setup & Installation — Simulation

Prerequisites

Build & Run

🚀 Setup & Installation — Real Robot

Prerequisites

Step 1 — Clone the Repository (on laptop)

Step 2 — Install Python Dependencies

Step 3 — Build ROS2 Workspace

Step 4 — Upload Arduino Code

Step 5 — Find Arduino Port on Raspberry Pi

Step 7 — Run the System

🌐 Web Dashboard

🛠️ Troubleshooting

✅ Project Status

Done

📋 To Do

📋 Project Info

About

Releases

Packages

Contributors

Languages

Name		Name	Last commit message	Last commit date
arduino code		arduino code
real robot		real robot
simulation		simulation
README.md		README.md

Latest commit

History

Folders and files

Repository files navigation

📌 Overview

🏆 Scoring System

🎯 How it works:

🏗️ System Architecture

🧠 Core Features

🗂️ Project Structure

🔧 Hardware Stack

🔌 Serial Connection

💻 Software Stack

📡 Serial Command Protocol

🏟️ Arena Setup

🚀 Setup & Installation — Simulation

Prerequisites

Build & Run

🚀 Setup & Installation — Real Robot

Prerequisites

Step 1 — Clone the Repository (on laptop)

Step 2 — Install Python Dependencies

Step 3 — Build ROS2 Workspace

Step 4 — Upload Arduino Code

Step 5 — Find Arduino Port on Raspberry Pi

Step 7 — Run the System

🌐 Web Dashboard

🛠️ Troubleshooting

✅ Project Status

Done

📋 To Do

📋 Project Info

About

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages