Computer science is also not hardware engineering. A computer science student has no idea how to get trains to report their position, because that requires hardware. For example, sensors laid along the track at reasonably closely spaced intervals which read an RFID transponder carried by the train, and send a message to some concentrator which forwards it to a data centre. It's okay not to have all the answers; transit systems are not a one person job.

A good answer to the question is "I don't really know the specific technology for sensing the positions of trains. If the sensing is used to make automatic driving decisions, it is of crucial importance that it be accurate and reliable, for safety reasons. If trains are automatic, then the system must not only know their positions, but the trains must know that the system knows their position; if a train's computer loses touch with the system, it must come to an emergency stop. A rogue train whose position isn't known must not continue moving, for obvious reasons. I think I would survey how all this has been successfully done in existing transit infrastructures around the world and borrow the best elements from the existing designs".