ROS-Noedic-ZED2-Wrapper-Docker

Files to create docker with ROS Melodic and ZED ROS Wrapper

To Build Docker Image:

1. chmod +x build-ros-desktop-image.sh
2. bash build-ros-desktop-image.sh

Create Docker Container

To Run Docker for the First Time:

1. xhost +si:localuser:root (only if it's not running on the robot)
2. docker run -it --runtime nvidia --privileged -e DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix 3.7-ros-gl-devel-cuda11.4-ubuntu20.04 (may add --net=host flag for communication between docker and jackal)

To Re-enter Docker Container After the First Time (with Existing container):

1. docker start (docker container's name)
2. docker exec -it (docker container's name) bash

Setup Docker Container:

Following procedure are required in every new terminal for docker container
Once Enter the Docker Container, you should see: root@(docker container ID):/opt/ros_ws#

Then Run: (in every terminal)

1. source devel/setup.bash

Open a New Terminal entering Jackal, Run:
2. ifconfig to obtain inet address under docker0 # For Example, 172.17.0.1
Then Go Back to Docker Container, Run:
3. export ROS_Master_URI=http://172.17.0.1:11311/ # Replace 172.17.0.1 with inet address

Run Our Code

To Run in Docker:

1. roslaunch zed_rtab_example zed_rtabmap.launch
2. Wait 5 seconds before moving camera for initialization
3. roslaunch zed_wrapper zed2.launch
4. cd /opt/ros_ws/mdp_cv/
5. python3 inf_class.py
6. cd /opt/ros_ws/cv-low-level
7. python3 hotdog.py

Record Rosbag

rosbag record rosout tf zed/mapPath zed/zed_node/pose zed/map zed/zed_node/rgb/image_rect_color cv_nav/road_percentage/forward cv_nav/road_percentage/rotateLeft cv_nav/road_percentage/rotateRight