🛠️ Autonomous Car Environment - ROS 2 + Python Project

📄 Project Description

This project is an educational exercise aimed at reinforcing knowledge of Python and ROS 2. It simulates an autonomous car navigating through cone-based tracks. The system is built with ROS 2 nodes, visualized with OpenCV, and structured for modular expansion.

---

🎯 Objective

The goal is to develop a new node or set of nodes (either in this package or in a new one) that allows the autonomous car to:

• Successfully navigate all available circuits autonomously.
• Use a single algorithm that works across all tracks (no per-track tuning allowed).
• Stop the car at the end of the circuit (near the finish area) before changing the state to "finish".

---

📡 Topic Interface & Communication Rules

✅ Developers can only interact with these topics:

Input topics to control the car:

Topic	Type	Description
/car/steering	std_msgs/Float32	Steering angle in radians. Positive = left, negative = right.
/car/acceleration	std_msgs/Float32	Linear acceleration in m/s². Positive = forward, negative = braking.
/car/state	std_msgs/String	Can only be set to "finish". Indicates completion of track.

⚠️ Limited access to /env topics

Topics starting with /env are internal and intended only for the environment's own use.

However, you can read from the topic:

• /env/car_position: shows the current position of the car (geometry_msgs/Point).

---

📊 Sensor-like Data Topics

Developers can read from these topics to simulate sensor data:

/can/* topics:

Topic	Type	Description
/can/speed	std_msgs/Float32	Current speed of the car.
/can/yaw	std_msgs/Float32	Current orientation (yaw).

/vision/* topics:

Topic	Type	Description
/vision/all_cones/blue	auria_msgs/PointArray	All blue cones (array of positions).
/vision/all_cones/yellow	auria_msgs/PointArray	All yellow cones.
/vision/nearby_cones/blue	auria_msgs/PointArray	Blue cones within 8 meters.
/vision/nearby_cones/yellow	auria_msgs/PointArray	Yellow cones within 8 meters.

Note: PointArray.msg is a custom message defined in the auria_msgs package:

geometry_msgs/Point[] points

---

🧠 Summary

Participants must write logic that:

• Consumes data from /can/* and /vision/* topics only.
• Publishes steering, acceleration, and state change commands via /car/*.
• Generalizes well to all tracks.
• Stops the car at the finish line before changing to "finish".

---

📘 ROS 2 Command Dictionary

🔎 Topic introspection

ros2 topic list                     # List all topics
ros2 topic info /car/steering       # Show info about a topic
ros2 topic type /can/yaw            # Show message type of topic
ros2 interface show std_msgs/msg/Float32  # Show message definition

👀 Topic reading (echo)

ros2 topic echo /can/yaw
ros2 topic echo /vision/all_cones/blue

📤 Publishing manually

ros2 topic pub /car/steering std_msgs/Float32 "{data: 0.5}" --rate 10
ros2 topic pub /car/acceleration std_msgs/Float32 "{data: 1.0}" --rate 10
ros2 topic pub /car/state std_msgs/String "{data: 'finish'}"

---

🐳 Docker Launch Instructions

To launch the environment with full GUI (rqt_graph, OpenCV, etc.), simply run the x script:

./x

---

📎 Developer Notes

• Run source install/setup.bash before launching any node, in each terminal.
• You may use rqt_graph to visualize connections.

---

🚀 Launch Instructions

▶️ Launch nodes individually

Make sure you have compiled and sourced your workspace:

colcon build --symlink-install
source install/setup.bash

Then run each node like this:

ros2 run environment env     # Launches the environment (map, cones, car view)
ros2 run environment car     # Launches the car node (publishes position)

🧩 Launch everything with one command

A launch file is provided for convenience. To run both the environment and car nodes together:

ros2 launch environment launch_env.py

This will start both nodes in the same terminal, printing their logs to the screen.

---

Happy coding & enjoy the challenge! 🚗🧠✨