rosbag2_2024_07_19_11_01_37 Dataset

# README for `rosbag2_2024_07_19_11_01_37` Dataset

## Overview

This dataset is a ROS 2 bag file (`rosbag2_2024_07_19-11_01_37.mcap`) containing recorded data from an autonomous vehicle system. It includes various sensor data (e.g., LiDAR point clouds, images, odometry, and IMU data) and system messages recorded during vehicle operations. The dataset is useful for research in autonomous driving, robotics, sensor fusion, or machine learning applications.

### Dataset Contents:

- **File Name**: `rosbag2_2024_07_19-11_01_37.mcap`

- **Size**: Varies based on the ROS 2 messages stored.

- **Storage Format**: `MCAP`

- **Total Duration**: Approximately 43.89 minutes (2633487071762 nanoseconds).

- **Total Message Count**: 53219 messages.

- **Total Topics**: 100+

## Data Structure

The dataset is organized into ROS 2 topics, each associated with a specific message type. These topics include data from various sensors, commands, vehicle state, and more.

### Topics Overview:

1. **/nn_cmd_vel_stamped** (148 messages): Contains velocity commands for navigation with time-stamped data using `geometry_msgs/msg/TwistStamped`.

2. **/CameraRear** (107 messages): Captures rear camera images using `sensor_msgs/msg/Image`.

3. **/lidar_safety/rear_left/cloud** (593 messages): Point cloud data from rear-left LiDAR using `sensor_msgs/msg/PointCloud2`.

4. **/imu/data** (627 messages): IMU data including orientation, angular velocity, and linear acceleration using `sensor_msgs/msg/Imu`.

5. **/odom_pub** (451 messages): Odometry data for vehicle position and velocity using `nav_msgs/msg/Odometry`.

6. **/cmd_vel** (message count: 0): Linear and angular velocity commands using `geometry_msgs/msg/Twist`.

7. **/joy** (1151 messages): Joystick data for manual control using `sensor_msgs/msg/Joy`.

For a complete list of topics, refer to the "topics_with_message_count" section of the dataset.

## Message Types:

- **geometry_msgs/msg/TwistStamped**: Stamped velocity commands for moving the vehicle.

- **sensor_msgs/msg/Image**: Images captured from cameras mounted on the vehicle.

- **sensor_msgs/msg/PointCloud2**: Point cloud data from LiDAR sensors.

- **nav_msgs/msg/Odometry**: Odometry information, including position, orientation, and velocity.

- **sensor_msgs/msg/Imu**: Inertial Measurement Unit (IMU) data for capturing orientation, velocity, and acceleration.

## Instructions for Use:

### Prerequisites:

- ROS 2 Humble or higher installed.

- A system capable of handling large data files.

### How to Load the Dataset:

1. **Download the dataset**: The dataset is available in MCAP format, which can be processed using ROS 2 tools.

2. **Play the Bag File**:

   ```bash

   ros2 bag play rosbag2_2024_07_19-11_01_37.mcap

   ```

   This command replays the recorded data, publishing the messages to the respective topics in real time. 

3. **View Topic Information**:

   You can check the list of available topics by running:

   ```bash

   ros2 bag info rosbag2_2024_07_19-11_01_37.mcap

   ```

   This provides a summary of topics and message counts, as well as additional metadata.

4. **Access Specific Data**:

   Use the `ros2 topic echo` command to access data from specific topics:

   ```bash

   ros2 topic echo /CameraRear

   ```

   Similarly, you can substitute the topic name to explore different types of data, such as LiDAR point clouds or IMU readings.

5. **Convert Data for Custom Use**:

   If you need to process specific messages (e.g., point clouds or images), you can use ROS 2 services to convert these into more accessible formats:

   - For images, use `ros2 bag export` or custom Python scripts to extract and save images.

   - For point cloud data, you may use tools like `pcl_ros` or custom scripts to export point cloud data for further analysis.

### Data Dictionary:

- **Timestamp**: Each message is associated with a precise timestamp, ensuring synchronized playback of the data.

- **QoS Profiles**: Various QoS (Quality of Service) profiles are used for different topics, allowing for flexibility in data communication (reliability, durability, etc.).

## Potential Applications:

- **Autonomous Driving Research**: Test perception algorithms, sensor fusion techniques, or navigation systems.

- **Machine Learning**: Use the dataset for training or validating machine learning models in object detection, scene understanding, and motion planning.

- **Robotics**: Simulate robot movement and interaction in controlled environments with realistic sensor data.

## Important Notes:

- Some topics may have zero messages; this is indicated in the "message_count" field.

- The dataset uses `MCAP` format, which offers improved storage efficiency and querying over previous formats like `SQLite3`.

If you have any questions or require further assistance, feel free to reach out to the dataset contributors or refer to the ROS 2 documentation.