ROSBag-Process

Reads, processes, and displays data contained within a ROSBag file

System Requirements

• Ubuntu 16.04 (native or via Docker)

Software Requirements

• ROS
• Python 2.7
• XQuartz (if you are running on Mac via Docker)

Required Python packages

Most of these can be installed with a simple 'pip install MODULE' command

• numpy
• opencv (pip install opencv-python)
• pyglet
• rosbag
• rospkg
• matplotlib (needs version 2.0+)
• cv_bridge (apt-get install ros-kinetic-cv-bridge)

KITTI to ROSbag conversion

Install Kitti2Bag

'pip install kitti2bag'

To convert, follow the directions on the Github page. Pay attention to the how the zip files are unpacked.

https://github.com/tomas789/kitti2bag

Command Line

Basic command:

python extract_rosbag.py <BAGFILE>

For visualizing only certain topics

python extract_rosbag.py --topics <TOPIC1>,<TOPIC2>,... <BAGFILE>

For extracting images

python extract_rosbag.py --outdir <OUTPUT_DIR> <BAGFILE>

To run via Docker on Mac:

• Install XQuartz running
• Make sure you have updated XQuartz settings for GLX

defaults write org.macosforge.xquartz.X11 enable_iglx -bool true

• Log Out and Log back Into Mac session
• Install socat on your Mac (brew install socat)

Then these steps once per Mac session:

• Run socat on your Mac host

socat TCP-LISTEN:6000,reuseaddr,fork UNIX-CLIENT:\"$DISPLAY\"

• Find out your host machine's IP address using 'ifconfig'
• Within your Docker container, setup your DISPLAY environment to point to your host's XQuartz

export DISPLAY=<HOST_IP_ADDRESS>:0