Border Patrol Bot

A ROS2-based autonomous border monitoring system that tracks objects crossing a defined boundary line.

Overview

Border Patrol Bot combines navigation and computer vision to create an autonomous security monitoring system. The robot patrols along a defined border line using SLAM and Nav2, while simultaneously detecting and tracking objects that cross the border using depth cameras, OpenCV, and object detection.

Features

• Autonomous Navigation: Uses slam_toolbox and Nav2 for SLAM and autonomous patrol
• Object Detection: YOLOv8 for real-time object detection and tracking
• Computer Vision: OpenCV for image processing and visualization
• Border Crossing Detection: Identifies objects crossing a configurable boundary line
• Entry/Exit Classification: Categorizes crossings as entries or exits based on robot position
• 3D Positioning: Uses point cloud data to locate objects in 3D space

Installation

# Clone the repository
git clone https://github.com/MuhammadBilalKhan267/Border-Patrol-Bot

# Build the package
colcon build --symlink-install

Usage

Launch Simulation

# Launch default simulation
ros2 launch patrol_bot launch_sim.launch.py

# Launch with custom world
ros2 launch patrol_bot launch_sim.launch.py world:=/home/bilal/robotics/ros2/patrol_bot/src/patrol_bot/worlds/obstacles.world

Start Border Crossing Detection

ros2 run crossing_detection trackerNode

Visualization

# Launch RViz with predefined configuration
rviz2 -d /home/bilal/robotics/ros2/patrol_bot/src/patrol_bot/config/slam.rviz

Manual Control

# Teleoperation using keyboard
ros2 run teleop_twist_keyboard teleop_twist_keyboard --ros-args -r cmd_vel:=cmd_vel_key

Monitoring Border Crossings

Border crossing data is published to the following topics:

• /border_crossing/counts: Summary statistics of border crossings
• /border_crossing/status: Detailed information about tracked objects

View the data in real-time:

ros2 topic echo /border_crossing/counts

Contributing

Contributions to Border Patrol Bot are welcome! Whether it's bug reports, feature requests, or code contributions, please feel free to reach out and help improve this project.