Jtc sum periods

[fmauch]

Sample trajectory based on the sum of periods instead of the absolute time.

This is basically the second part of #1191. In the long run I want to be able to scale the period with a scaling factor in order to artificially slow down / speed up progression in the trajectory. This PR is there to separate the necessary time counting change from the scaling functionality to make things more contained and easier to review.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 83.57%. Comparing base (d8b30f8) to head (9e97d9f).
Report is 1 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1395      +/-   ##
==========================================
+ Coverage   83.56%   83.57%   +0.01%     
==========================================
  Files         122      122              
  Lines       10987    10991       +4     
  Branches      937      936       -1     
==========================================
+ Hits         9181     9186       +5     
  Misses       1492     1492              
+ Partials      314      313       -1     

Flag	Coverage Δ
unittests	83.57% <100.00%> (+0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
...jectory_controller/joint_trajectory_controller.hpp	100.00% <ø> (ø)
...ory_controller/src/joint_trajectory_controller.cpp	83.90% <100.00%> (+0.04%)	⬆️
...ectory_controller/test/test_trajectory_actions.cpp	97.76% <100.00%> (+0.01%)	⬆️

... and 3 files with indirect coverage changes

[christophfroehlich]

LGTM

[saikishor]

Looks good to me.

Doing this way, we are kinda doing it open loop with the time. Am I right?

[fmauch]

Looks good to me.

Doing this way, we are kinda doing it open loop with the time. Am I right?

Yes, that is true. That requires correct period times without missing or skipping a period. Which is why we recently found issues with that and fixed them. From what I can draw from the design concepts this is an assumption we can make here, though.

[saikishor]

Yes, that is true. That requires correct period times without missing or skipping a period. Which is why we recently found issues with that and fixed them. From what I can draw from the design concepts this is an assumption we can make here, though.

Got it. Thanks for the explanation.

[fmauch]

Semi-binary builds should now also succeed, now that ros-controls/ros2_control#1898 is merged. I cannot restart the checks, though.