dddmr_perception_3d

Perception 3D is graph-based framework allowing user to develope applications for mobile robots, such as path planning, marking/clearing obstacles, creating no-enter/speed limit layer. You can reference:

• dddmr_global_planner
• dddmr_local_planner

Global planning in 3D map

Marking/Tracking/Clearing

Speed-limit/no-enter zone

Perception 3D:

• Sensor support:
  • Multilayer spinning lidar (Velodyne/Ouster/Leishen)
  • Depth camera (Realsense/oak)
  • Scanning Lidar (Livox mid-360/Unitree 4D LiDAR L1)
• Zone feature support:
  • Static layer
  • Speed limit layer
  • No enter layer

Multilayer Lidar Demo (Leishen Lidar C16)

Click me to see tutorial

1. Create docker image

The package runs in the docker, so we need to build the image first. We support both x64 (tested in intel NUC) and arm64 (tested in nvidia jetson jpack5.1.3/6).

cd ~
git clone https://github.com/dddmobilerobot/dddmr_navigation.git
cd ~/dddmr_navigation && git submodule init && git submodule update
cd ~/dddmr_navigation/dddmr_docker/docker_file && ./build.bash

2. Download essential files

ROS2 bag that contains multilayer lidar from Leishen C16 will be download to run the demo.

cd ~/dddmr_navigation/src/dddmr_perception_3d && ./download_files.bash

3. Run demo

Create a docker container

[!NOTE] The following command will create an interactive docker container using the image we built. We will launch the demo manually in the container.

cd ~/dddmr_navigation/dddmr_docker && ./run_demo.bash

Launch everything in the container

The bag file will be auto-played after 5 seconds when launching.

cd ~/dddmr_navigation && source /opt/ros/humble/setup.bash && colcon build --symlink-install --cmake-args -DCMAKE_BUILD_TYPE=Release
source install/setup.bash
ros2 launch perception_3d multilayer_spinning_lidar_3d_ros_launch.py

Multiple Depth Cameras Demo (Realsense D455)

Click me to see tutorial

1. Create docker image

The package runs in the docker, so we need to build the image first. We support both x64 (tested in intel NUC) and arm64 (tested in nvidia jetson jpack5.1.3/6).

cd ~
git clone https://github.com/dddmobilerobot/dddmr_navigation.git
cd ~/dddmr_navigation && git submodule init && git submodule update
cd ~/dddmr_navigation/dddmr_docker/docker_file && ./build.bash

2. Download essential files

ROS2 bag that contains depth images from two cameras will be download to run the demo.

cd ~/dddmr_navigation/src/dddmr_perception_3d && ./download_files.bash

3. Run demo

Create a docker container

[!NOTE] The following command will create an interactive docker container using the image we built. We will launch the demo manually in the container.

cd ~/dddmr_navigation/dddmr_docker && ./run_demo.bash

Launch everything in the container

The bag file will be auto-played after 5 seconds when launching.

cd ~/dddmr_navigation && source /opt/ros/humble/setup.bash && colcon build --symlink-install --cmake-args -DCMAKE_BUILD_TYPE=Release
source install/setup.bash
ros2 launch perception_3d multi_depth_camera_3d_ros_launch.py