Line Following with Agent-Based Decision Making

[akrambzeo]

Implemented an agent for decision making in line following, added a trajectory buffer and networks, and created a plot file for analysis of the results.

[BlueAndi]

Please increase number of PRs and keep logical changes small.

[gabryelreyes]

For every python function that returns something, please try to use type-hinting for the return values on the function definition. This helps better follow and understand the code

[gabryelreyes]

This works. But could be done better by using if-else blocks

[hoeftjch]

Reinforcement training loop runs as expected. Further architecture changes needed to drastically reduce coupling between environment and RL Agent. This will be done in future updates regarding issue #156.