You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thanks to @MegMll, we now have benchmarks that show a performance increase up to 100x faster when using rclcpp (ros2 in c++) for sending eigen matrix over a topic, rather than using rclpy (ros2 for python).
Thus, we should move this part of the code to c++.
I think the interpolator node can be easily completely re-written in c++, but for the MPC node, if it is too annoying, we could just write a c++ lib that manage the ros messages and bind it to python...
Cheers
The text was updated successfully, but these errors were encountered:
Hello,
Thanks to @MegMll, we now have benchmarks that show a performance increase up to 100x faster when using rclcpp (ros2 in c++) for sending eigen matrix over a topic, rather than using rclpy (ros2 for python).
(The benchmark in question https://github.com/MegMll/timings_ros2_conversion_msg)
Thus, we should move this part of the code to c++.
I think the interpolator node can be easily completely re-written in c++, but for the MPC node, if it is too annoying, we could just write a c++ lib that manage the ros messages and bind it to python...
Cheers
The text was updated successfully, but these errors were encountered: