ROS2 Depth to Point Cloud Converter

This project demonstrates depth image to point cloud conversion in a differential drive robot simulation using ros_gz_bridge. The package processes depth camera data and converts it to point cloud format using ros_gz_bridge for real-time 3D visualization and perception applications. View the project visualization here.

System Requirements

• Ubuntu: 22.04 LTS
• ROS 2: Jazzy Jalisco
• Gazebo: Harmonic

Usage

Prerequisites

Install required ROS 2 packages:

sudo apt install ros-jazzy-robot-state-publisher \
                 ros-jazzy-joint-state-publisher \
                 ros-jazzy-xacro \
                 ros-jazzy-teleop-twist-keyboard \
                 ros-jazzy-ros-gz-sim \
                 ros-jazzy-ros-gz-bridge \
                 ros-jazzy-sensor-msgs \
                 ros-jazzy-rviz2

Installation & Setup

1. Create Workspace and Clone Repository

mkdir -p my_ws/src && cd my_ws/src
git clone https://github.com/yashikasharma0301/depth_to_pcl.git

2. Build the Workspace

cd ..
colcon build

3. Source the Workspace

source install/setup.bash

Running the Simulation

1. Launch Robot with Depth Camera in Gazebo

ros2 launch depth_to_pcl gazebo_spawn.launch.py

To enable robot control within Gazebo:

• Click on the three dots on the upper right corner of your Gazebo window
• Search for Teleop from the menu and click on it
• This will open the teleop plugin allowing you to control the robot directly from Gazebo

2. Start RViz2 for Point Cloud Visualization

In another terminal run:

rviz2 -d /home/yashika-arz-i007/my_ws/src/depth_to_pcl/rviz/pcl.rviz

• Ensure the Fixed Frame is set to camera_optical_frame
• You will be able to see the point cloud visualization in RViz2
• Move the robot around in Gazebo using Teleop and you will be able to see the corresponding point cloud

3. Alternative: Depth Image Visualization

If you want to see the depth image instead of the point cloud:

• Turn off the point cloud display in RViz2
• Set the Fixed Frame to base_footprint
• Select the corresponding RViz plugins for camera and image visualization
• You will be able to see some rough visualization in the set format (Use rqt image viewer for better visualization)

Credits & References

Robot Model: This project uses a URDF model adapted from the TortoiseBot example in the OSRF ROS Book. The original model has been modified for ROS 2 Jazzy and Gazebo Harmonic integration with depth camera sensor capabilities.

Original Authors: Open Source Robotics Foundation (OSRF)