[Simulator]: Repititive use of target changer is not possible

Summary

In my example scenario, the target changer on the left hand side should redirect agents to the target on the right hand side. The target on the right hand side should do the opposite. Goal: agents walk from right to left, from left to right and so on ("infinite loop"). (proceedAgents_problem)

What is the current bug behavior?

Currently, each target changer is only used once. As a result, the single agent in the movie above has the targets [1,2,1]. I do not have an "infinite loop".

What is the expected correct behavior?

The targets of the agent should be [1,2,1, ...,2,1,2 ,... ] which is set iteratively by the target changers.

Starting point

The reason why each target changer is only used once is the definition of the update method of the TargetChangerController.class. Here, processedAgents.containsKey(agent.getId()) is only true for the first time.

Is this behavior really required? Why?