ROSbot 2R Navigation Project

Overview

This project explores autonomous navigation using the ROSbot 2R, implementing two distinct path-planning approaches to efficiently reach a target destination while avoiding obstacles. The goal was to measure total travel time and distance covered while ensuring smooth navigation in both simulated and real-world environments.

Navigation Approaches 🚀

1️⃣ Multi-Waypoint Navigation

• The robot follows a predefined set of waypoints, optimizing the path to reach the destination efficiently.
• This approach enables flexible route planning and smooth movement across multiple checkpoints.
• It is particularly effective for structured environments where obstacle placement is predictable.

2️⃣ Single-Coordinate Navigation

• The robot directly moves toward a single target coordinate while actively avoiding obstacles.
• Suitable for dynamic environments where predefining waypoints is impractical.
• Provides a more adaptive response to unexpected obstacles and changes in surroundings.

Implementation 🔧

The project was implemented in simulation using Gazebo and RViz, followed by real-world deployment on the ROSbot 2R.

🔍 Sensor Suite

To achieve robust navigation, the robot utilized:

• LiDAR → Distance measuring and obstacle detection.
• Time-of-Flight (ToF) Sensors (FL & FR) → Short-range obstacle avoidance.
• IMU → Orientation and motion tracking.
• Wheel Encoders → Precise odometry data.
• Camera → Used for visualization purposes only.

🎯 Simulation Results

• Both algorithms were successfully tested in Gazebo.
• The robot consistently reached the goal without collisions, efficiently selecting routes based on sensor data.
• Waypoint-based navigation proved effective in structured layouts, while single-coordinate navigation adapted better to dynamic environments.

📹 Real-World Testing

• The system was successfully transferred to a physical ROSbot 2R.
• Real-world performance aligned closely with simulation, demonstrating robust obstacle avoidance and smooth navigation.
• The results validated the sensor fusion and path-planning techniques implemented.

Key Takeaways 🎓

• Multi-waypoint navigation is better for predefined paths, while single-coordinate navigation offers adaptability.
• Simulation closely matched real-world results, showcasing the reliability of the approach.
• Sensor fusion played a crucial role in precise movement and real-time decision-making.

---

💡 Next Steps: Expanding the project to incorporate dynamic obstacle avoidance using deep learning and refining real-world implementation.

🚀 Stay tuned for updates!