ros_canopen
ros_canopen copied to clipboard
ros-industrial
socketcan_bridge support for DBC files
Add encoding/decoding step to the socketcan_bridge:
- Load DBC Files that contain Frame/Signal specification (http://vector.com/vi_candb_en.html)
- Create pub and sub for DecodedCANFrames
This is a feature request. If somebody is interested, I can help developing this feature.
-Rein
On Wed, Feb 17, 2016 at 10:49 AM, Mathias Lüdtke [email protected] wrote:
Do you plan to work on this or is it just a feature request?
— Reply to this email directly or view it on GitHub https://github.com/ros-industrial/ros_canopen/issues/160#issuecomment-185123683 .
I don't really see how you envision the DecodedCANFrames message to work or what it benefits will be. Of course, for some CAN messages such a conversion node would be useful for example CAN enabled encoders could be conveniently converted to sensor_msgs/JointState. I guess you could make a node which converts specific CAN messages to specific ROS messages. Should be an additional package in my opinion though.
I have done something similar for canopen_chain_node: each object can be published as a single topic with one of the std_msgs types.
@ipa-mdl,This indeed could be an option; however then you loose the relation between signals of the same CAN Frame. @iwanders, you might be right, maybe a more nice way is developing an abstract base class with a decodedframe callback that maps the decoded can frames on ROS messages ...
If you really want to publishe complex messages then you should go for a python-can implementation or a converter node that listens to the socketcan_bridge output.
I don't think support for DBC files should be added to socketcan_bridge package.
Example: https://github.com/NewEagleRaptor/pacifica-dbw-ros
dbc is a package for supporting DBC files
dbw_pacifica_can is a package that reads/writes CAN messages from/to socketcan_bridge topics using the above dbc library
It's really hard to generalize this implementation IMHO.
Hello Mike, Suppose I have a .dbc file and if I need to visualize the data in a human readable format, what we shall do? Do we have some documentation for proceeding further?
