EXTRACT ROSBAG

Extracts a rosbag file (ROS) and converts it to a EUROC/ASL format.

The basic structure of the format is a set of directories containing:

├── robot0
   ├── camera
   │    ├── /*.png
   ├── gps0
   │    ├── /data.csv
   ├── gps_filtered
   │     ├── /data.csv
   ├── ground_truth
   │    ├── /data.csv
   ├── imu0
   │    ├── /data.csv
   ├── lidar
   │    ├── /*.pcd
   ├──odom
   │    ├── /data.csv
   ├── odometry_gps
   │    ├── /data.csv  
   ├──tf
   │    ├── /data.csv
   └── tf_static
        ├── /data.csv

INSTALL

Install the requirements with:

• ./install.sh

USAGE

First we activate the virtual environment created during installation.

• source setup.sh

Now, we just run the python file once the YAML file was properly configured.

• python3 extract_rosbag.py

Setup YAML

Configure the following variables in config.yaml, for example:

• rosbag_path: "/home/user/file.bag"

• output_path: "/home/user/extract_rosbag"

Next: configure the topics that will be read and extracted

topic_name_point_cloud: "/ouster/points" # /os1/pointCloud
topic_name_odometry: "/odometry/filtered"
topic_name_ground_truth: "/ground_truth/state"
topic_name_gps: "/fix"
topic_name_imu: "/imu/data"
topic_name_camera: "/aravis_cam/aravis_cam/image_raw"
topic_name_tf: "/tf"
topic_name_tf_static: "/tf_static"
topic_name_gps_filtered: "/gps/filtered"
topic_name_odometry_gps: "/odometry/gps"\

Finally, decide whether to extract those topics:
extract_topics:
odometry: True
lidar2pcd: True
lidar2csv: False
ground_truth: False #just for simulation
gps: True
imu: True
camera: False
tf: True
tf_static: True
gps_filtered: True
odometry_gps: True\