nav_core2

[reinzor]

Are you planning to be compatible with https://github.com/locusrobotics/robot_navigation/tree/master/nav_core2 ?

[corot]

Not by now, but we support nav_core-based pluggins, so I suppose we can load too nav_core2 plugins using the https://github.com/locusrobotics/robot_navigation/tree/master/nav_core_adapter package.

[reinzor]

Thanks

…

-Rein

On Wed, Nov 22, 2017 at 10:48 PM, Jorge Santos Simón < ***@***.***> wrote: Not by now, but we support nav_core-based pluggins, so I suppose we can load too nav_core2 plugins using the https://github.com/ locusrobotics/robot_navigation/tree/master/nav_core_adapter package. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#11 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AD-4l4eBgYMwUhhLp_P7gB6dUWEpcjv4ks5s5Ja9gaJpZM4Ql4br> .

[reinzor]

Thanks for your answer @corot. I have been looking at the abstraction you have created for the controller and the planner (https://github.com/magazino/move_base_flex/blob/master/move_base_flex_core/include/move_base_flex_core/abstract_local_planner.h https://github.com/magazino/move_base_flex/blob/master/move_base_flex_core/include/move_base_flex_core/abstract_global_planner.h). I also looked at the ones defined here in the nav_core2 interface (https://github.com/locusrobotics/robot_navigation/blob/master/nav_core2/include/nav_core2/global_planner.h https://github.com/locusrobotics/robot_navigation/blob/master/nav_core2/include/nav_core2/local_planner.h).

I think the interfaces you defined are nice, except for the interfaces in the local planner for computing the velocity and the isGoalReached method. Because the state used in this method is not an input argument but magically obtained via the TF listener or an odom subscriber. In the nav_core2 interface, this has been changed. What do you think about this?

[corot]

I agree; #17

[reinzor]

Okay, I will close this issue