Autonomous Object Detection and Navigation System with TurtleBot3 Burger in ROS2

​

Our project aims to enhance the performance of the TurtleBot3 Burger by implementing an autonomous object detection and navigation system. To achieve this, we utilize a personal laptop as the central processing unit, running YOLOv8 for efficient object detection. The initial challenge we addressed was the high latency of up to 1 second, which we successfully mitigated through a novel approach. Additionally, our system is seamlessly integrated with ROS2, providing a robust and modular framework for robot control and communication.

Key Components:

​

1. Hardware:

   • TurtleBot3 Burger: Our robotic platform of choice, equipped with sensors and actuators, forms the physical base for our autonomous system.

   • Personal Laptop: Serving as the brain of the operation, the laptop runs the YOLOv8 object detection model, ROS2 middleware, and controls the decision-making process.

2. Object Detection:

   • YOLOv8 Implementation: We leverage the capabilities of YOLOv8 for real-time object detection. This advanced model allows us to accurately identify and classify objects in the robot's environment.

3. ROS2 Integration:

   • Communication Framework: Our system is integrated with ROS2, a powerful middleware for robot control and communication. This integration enhances the flexibility and scalability of the system, allowing for seamless interaction between different components.

4. Latency Reduction:

   • Data Compression: To address the initial latency challenge, we implemented a data compression mechanism. The TurtleBot3 captures data, compresses it, and sends a compressed version to the laptop. On the laptop side, the data is uncompressed, significantly reducing latency.

5. Human Detection and Navigation:

   • Custom Package: We developed a specialized ROS2 package that converts human detection data into command velocity (cmd_vel). This package enables the robot to respond to the presence of humans in its vicinity, facilitating safe and adaptive navigation.

6. Control System:

   • PID Controller: Our system incorporates a Proportional-Integral-Derivative (PID) controller, enhanced by ROS2 middleware. This combination ensures the precision of the robot's movements, providing smoother navigation and better responsiveness to environmental stimuli.