Codecov Report

Merging #763 (6769fe8) into main (48625f6) will increase coverage by 0.69%.
The diff coverage is n/a.

❗ Current head 6769fe8 differs from pull request most recent head 24f8abd. Consider uploading reports for the commit 24f8abd to get more accurate results

@@            Coverage Diff             @@
##             main     #763      +/-   ##
==========================================
+ Coverage   54.87%   55.56%   +0.69%     
==========================================
  Files         196      198       +2     
  Lines       21358    21420      +62     
==========================================
+ Hits        11719    11900     +181     
+ Misses       9639     9520     -119     

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 48625f6...24f8abd. Read the comment docs.

I am really confused as to how this would affect the servo tests in any way. None of the servo PRs seem to be failing but this one is.

I am having the following issues with this branch:

• Multiple consecutive requests are not possible. In single_plan_execution.cpp the call_once of the local planner ensures that the local planner is only executed once. However, this is not reset when the local planner was successful. Replacing the std::once_flag with a Boolean and resetting after stopping the local planner seems to solve this. However I'm not sure this is thread safe.
• When the global planner does not find a solution (Not possible or collision after smoothing) this crashes the Hybrid Planner Component container. In other words: if the initial path is not possible, it will not try again, or return that this path is not possible, but the hybrid container will stop.
• It is not possible to pass a ompl_planning.yaml file to change the ompl planning parameters for the global planning.
• The following parameters in the MotionPlanRequest are ignored:
  • max_velocity_scaling_factor
  • max_acceleration_scaling_factor
  • max_cartesian_speed
  • workspace_parameters.*
  • start_state.*

I find this branch has broken the demo. One of the reasons, at least, is that the trajectory controller does not come up.

I am having the following issues with this branch:

• Multiple consecutive requests are not possible. In single_plan_execution.cpp the call_once of the local planner ensures that the local planner is only executed once. However, this is not reset when the local planner was successful. Replacing the std::once_flag with a Boolean and resetting after stopping the local planner seems to solve this. However I'm not sure this is thread safe.

• When the global planner does not find a solution (Not possible or collision after smoothing) this crashes the Hybrid Planner Component container. In other words: if the initial path is not possible, it will not try again, or return that this path is not possible, but the hybrid container will stop.

• It is not possible to pass a ompl_planning.yaml file to change the ompl planning parameters for the global planning.

• The following parameters in the MotionPlanRequest are ignored:

  • max_velocity_scaling_factor
  • max_acceleration_scaling_factor
  • max_cartesian_speed
  • workspace_parameters.*
  • start_state.*

@JensVanhooydonck Hello, right now I also try to use hybrid planning. Passing ompl_planning.yaml and modify max_velocity_scaling_factor do not work as you described. Have you figure out how to fix it? Any advice will be highly appreciated. Thanks!

I am having the following issues with this branch:

• Multiple consecutive requests are not possible. In single_plan_execution.cpp the call_once of the local planner ensures that the local planner is only executed once. However, this is not reset when the local planner was successful. Replacing the std::once_flag with a Boolean and resetting after stopping the local planner seems to solve this. However I'm not sure this is thread safe.

• When the global planner does not find a solution (Not possible or collision after smoothing) this crashes the Hybrid Planner Component container. In other words: if the initial path is not possible, it will not try again, or return that this path is not possible, but the hybrid container will stop.

• It is not possible to pass a ompl_planning.yaml file to change the ompl planning parameters for the global planning.

• The following parameters in the MotionPlanRequest are ignored:

  • max_velocity_scaling_factor
  • max_acceleration_scaling_factor
  • max_cartesian_speed
  • workspace_parameters.*
  • start_state.*

@JensVanhooydonck Hello, right now I also try to use hybrid planning. Passing ompl_planning.yaml and modify max_velocity_scaling_factor do not work as you described. Have you figure out how to fix it? Any advice will be highly appreciated. Thanks!

I didn't try this in the last year.. I don't know what the status is currently..

@JensVanhooydonck Thanks for reply! I will try to find a solution.