Refining Robotics Academy exercises

The goal of this week was to help in the refinement of the exercises migrated to Ros2 in Robotics Academy.

Changes to all of the exercises

I changed how the interfaces for the Hardware Abstraction Layered are imported from using the original local interfaces to the ones shared globally in the python package hal_interfaces in order to save space and make the exercises simpler.

Other major modifications were made with the changes to the method used to retrieve data from the sensors, going from an always spinning method or an in-request spin to an hybrid approach using threading and sleeps to perform the spin at a fixed interval.

Specific changes to Obstacle Avoidance

In this exercise I have changed the UI to remove the useless camera canvas that was not needed as this view was already shown in the simulation below. And as we didn’t need the camera anymore I removed it from the car model in order to gain a 15% improved RTF and a 25% improvement in FPS.

All of the PR for the changes mentioned above can be found here.