This is the source code for round 1 - simulation car.

• Repository for car system: https://github.com/vietanhdev/autonomous-car-2020.
• Repository for road segmentation: https://github.com/vietanhdev/autonomous-car-2020-road-seg.
• Repository for traffic sign detection: https://github.com/vietanhdev/autonomous-car-2020-sign-detection.
• Previous version (2018 - with Watershed, Flood fill and HOG_SVM): https://github.com/vietanhdev/autonomous-car-2018.
• Resources: https://1drv.ms/u/s!Av71xxzl6mYZgdZp4HaKuMV4vIzqcA?e=eHUjyr.

• Team name: teamict

• Ubuntu 18.04
• Miniconda or Anaconda
• OpenCV 3.4.3: <scripts/install_opencv_3.sh>

• Install ROS Melodic - full desktop version: http://wiki.ros.org/melodic/Installation/Ubuntu

We need to initialize catkin workspace at the first time (build folders for projects).

cd main_ws
catkin_make

Add workspace to PATH:

echo "source !(pwd)/main_ws/devel/setup.bash" >> ~/.bashrc
source ~/.bashrc

cd main_ws
conda env create -n cds -f environment.yml 
conda activate cds

• rosbridge-suite

sudo apt-get install ros-melodic-rosbridge-server

cd  main_ws
catkin_make

NOTE: Don't use conda environment here!!!

roslaunch teamict server.launch

Enter Team name: teamict and server address: ws://localhost:9090.

Open another Terminal and type:

conda activate cds
roslaunch teamict teamict.launch

OR

conda activate cds
rosrun teamict teamict_node.py

Because we could not use cv2.imshow() to show image in this environment, <main_ws/src/teamict/src/debug_stream.py> is developed to view debug images.

Please use this command to open debug image viewer:

rosrun rqt_image_view rqt_image_view

• Source code for running car (we will mainly work here): <main_ws/src/teamict/src>.
• In that folder:
  • teamict_node.py: main node file
  • config.py: configuration file
  • debug_stream.py: stream of debug images. Please see the usage in teamict_node.py
  • image_processor.py: where to receive images and process.

• Pass debug_stream instance created in teamict_node.py to any class you want to debug.
• Create a stream for image using: debug_stream.create_stream('<stream_name>', '<topic_path>')

debug_stream.create_stream('depth', 'debug/depth')

• Publish image: debug_stream.update_image('<stream_name>', <image>)

debug_stream.update_image('depth', depth_image)

• Open debug image viewer:

rosrun rqt_image_view rqt_image_view

mkdir video_rgb
cd video_rgb
rosrun image_view image_saver image:=/teamict/camera/rgb _image_transport:=compressed

mkdir video_depth
cd video_depth
rosrun image_view image_saver image:=/teamict/camera/depth _image_transport:=compressed

In video_rgb:

ffmpeg -framerate 25 -i left%04d.jpg -c:v libx264 -profile:v high -crf 20 -pix_fmt yuv420p 01_rgb.mp4

In video_depth:

ffmpeg -framerate 25 -i left%04d.jpg -c:v libx264 -profile:v high -crf 20 -pix_fmt yuv420p 01_depth.mp4

• https://answers.ros.org/question/220502/image-subscriber-lag-despite-queue-1/