ifl-camp / tf_bag Goto Github PK

We have been using this tool for some per-processing and I have had a few issues with the TF tree update not getting triggered.

The tree looks something like
world -> gps_frame -> base_link -> point_cloud

world -> gps_frame is 10hz and gps_frame ->base_link is 1hz.

When using lookup transform from (world -> pointcloud) the TF is only updated at 1Hz. By splitting it into (world -> gps_frame) and (gps_frame -> point_cloud) it updates at 10hz.

Have had a look at code and nothing stands out as issue but have also had some difficulty with a lookupTransformWhenTransformUpdates and setting triggers so have resorted to similar solution but now wondering if its the TF tree or some other issue.

I'm sorry for the simple question, but I am not able to import this package.

I tried to install the package with
pip install -e git+https://github.com/IFL-CAMP/tf_bag#egg=tf_bag

and got this message

Obtaining tf_bag from git+https://github.com/IFL-CAMP/tf_bag#egg=tf_bag
  Cloning https://github.com/IFL-CAMP/tf_bag to ./src/tf-bag
    Complete output from command python setup.py egg_info:
    running egg_info
    creating src/tf_bag.egg-info
    writing src/tf_bag.egg-info/PKG-INFO
    writing top-level names to src/tf_bag.egg-info/top_level.txt
    writing dependency_links to src/tf_bag.egg-info/dependency_links.txt
    writing manifest file 'src/tf_bag.egg-info/SOURCES.txt'
    error: package directory 'src/tf_bag' does not exist
    
    ----------------------------------------
Command "python setup.py egg_info" failed with error code 1 in /home/sarah/practice_ws/src/sim_eval/sim_eval/src/tf-bag/

Am I installing it incorrectly?
Any help would be appreciated.
Thank you in advance.

Hi,

thank you for this package!
Just I would like to point out the misspelling in documentation where you say:
from bag_tf import BagTfTransformer and it should be : from tf_bag import BagTfTransformer

Hello!

First of all, thank you very much for the awesome package; it saved me a ton of headache!

I had some trouble loading up the messages from my bag (recorded in ROS Indigo) due to the line

self.tf_messages = sorted((tm for m in bag if m.topic == '/tf' for tm in m.message.transforms),
key=lambda tfm: tfm.header.stamp.to_nsec())

EDIT: I recorded my bag by running rosbag record odom tf.

(which can be found here in your code)

This only allows for messages under the "/tf" topic. However, for whatever reason, my bag's topic is "tf" and not "/tf". I'm not sure whether I made some mistake when recording it, but the fact is that changing your code to "tf" fixed the issue, and I can read the info I need from the bag effortlessly.

Is this some sort of mismatch in ROS versions? Should this be accounted for in the package? If you think so, I'd be happy to submit a pull request that implements compatibility with both topic names.