feat(53): Panoptic Segmentation Tests

[okrusch]

Description

I worked on the EfficientPS Network and tried to make it work. After a lot of hours of debugging i decided to intergrate a new Segmentation model and build a quick and dirty rosnode. It can implement several pretrained pyTorch models.

Fixes # (53)

Type of change

Please delete options that are not relevant.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)

Does this PR introduce a breaking change?

Not yet, it comes more or less with testing phases-

Most important changes

Switched out EfficientPs with differnet Segementation Models.

Checklist:

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works (might be obsolete with CI later on)
• New and existing unit tests pass locally with my changes (might be obsolete with CI later on)

[github-actions]

Simulation results

Metric	Value
Avg. driving score	0.0
Avg. route completion	0.0
Avg. infraction penalty	0.7
Collisions with pedestrians	0.7
Collisions with vehicles	0.0
Collisions with layout	0.0
Red lights infractions	0.0
Stop sign infractions	0.0
Off-road infractions	0.0
Route deviations	0
Route timeouts	0.0
Agent blocked	2000.0
Yield emergency vehicles infractions	0.0
Scenario timeouts	0.0
Min speed infractions	2000.0
undefined	0.0

[JuliusMiller]

Nice work!
Should we merge the testing directly into the main branch?
Or wait until the testing has finished?