I am modeling a software that requires to be interfaced with an optical tracking sensor, to measure the pose of some objects through optical markers. A general user interacts with this sensor (say External System 1) in two different use cases (say UC1 and UC2). I would like to model also the possibility that an admin user interacts with the software through the same use cases UC1 and UC2, but he can be interfaced with a software simulator (say External System 2) that simulates the real optical sensor.
I'm trying to model this scenario in a Use-Case diagram, but I am uncertain about the proper modeling of the presence of the simulator and the admin role. I wouldn't create a second pair of Use Case, because they are supposed to be exactly the same adopted by the general user. With reference to the Figure, I thought to add some dependency arrows (dashed) from the UCs to the External System 2 also, but this way would sound as also the general user could interact with the simulator, while this should not be allowed.
I apologize if this question could appear to be quite trivial, but these are my first experiences with software UML modeling.
Thank you in advance for your help.
Start with looking at your basic system like this
So basically your system is measuring something using a sensor. This Sensor
can be either a real sensor or a simulator. User
must not really care (though he knows what is attached). But the Measuring System will really not (!) care. So the simulator must simulate whatever is needed so the system will think it's a normal sensor.
Now that simulator is another system:
From the perspective of that system you look at Measuring System as external actor. Plus it offers an admin the possibility to set parameters for the test cycle.