[ros-users] Another question about policy based synchronizer

[Ken Conley]

Hi Aditya,

The sequence number is not a user-settable field. We hope to make it
user-settable in the future, but in our previous attempt to do so, the
feedback from ros-users was not to.

HTH,
Ken

On Mon, Oct 18, 2010 at 9:28 AM, Aditya Gadre <agadre at vt.edu> wrote:
> I have a working test code for policy based (approximate time) synchronizer
> in which one node publishes two data messages and the other node runs a
> synchronizer. One message is published at 1Hz and the other at 0.5Hz. For
> each message, I increment the relevant sequence number in the header.
> However, when these messages are received by the node running the policy
> based synchronizer, the sequence numbers are off by 1, although the time
> stamps match perfectly.
>
> For example, following output shows a simple run -
>
> Publisher log:
> [ INFO] [1287418571.114515026]: Starting test_policy_synchronizer_publisher
> node.
> [ INFO] [1287418571.122496486]: Publishing AHRS data: time =
> 1287418571.122447, seq = 1
> [ INFO] [1287418572.122643642]: Publishing AHRS data: time =
> 1287418572.122516, seq = 2
> [ INFO] [1287418572.122956042]: Publishing Alasca data: time =
> 1287418572.122883, seq = 1
> [ INFO] [1287418573.122685477]: Publishing AHRS data: time =
> 1287418573.122559, seq = 3
> [ INFO] [1287418574.122935718]: Publishing AHRS data: time =
> 1287418574.122809, seq = 4
> [ INFO] [1287418574.123219204]: Publishing Alasca data: time =
> 1287418574.123158, seq = 2
> [ INFO] [1287418581.122753362]: Shutting down
> test_policy_synchronizer_publisher node.
>
> Synchronizer node log:
> [ INFO] [1287418573.123472169]: ----------
> [ INFO] [1287418573.123744969]: Received message set
> [ INFO] [1287418573.123947997]: AHRS: timestamp = 1287418572.122516, seq = 1
> [ INFO] [1287418573.124106258]: Alasca: timestamp = 1287418572.122883, seq =
> 0
> [ INFO] [1287418575.122839197]: ----------
> [ INFO] [1287418575.122900518]: Received message set
> [ INFO] [1287418575.122951782]: AHRS: timestamp = 1287418574.122809, seq = 3
> [ INFO] [1287418575.123003883]: Alasca: timestamp = 1287418574.123158, seq =
> 1
>
> Comparing time stamps of sent and received messages, I see that the sequence
> numbers in the synchronizer node are off by 1 (if sent message seq = 3 then
> received message seq = 2). I am not sure why this discrepancy is popping up.
> I would appreciate any help in this regard! I have attached source code for
> both nodes.
>
> Thank you.
> - Aditya
>
> _______________________________________________
> ros-users mailing list
> ros-users at code.ros.org
> https://code.ros.org/mailman/listinfo/ros-users
>
>