Hi,
the question is about hobby electrical trains. First please take a look at the block diagram and the parts. The situation occures when a train pass from the block controlled by dev1 to the block controlled by the dev2. Both motor drives are driven by the same signal, but the "polarity" or phase of the motordrives' output may be not the same for a certain reason. The dev2 has the abillity to invert the common signal. When the two blocks are bridged and the signal in dev2 is inverted then the buck converters are in series and short circuited. In the dev2 a comparator compares the v2 to the let's say 2/3 of the nominal 16v, the output of the comparator is connected to the microcontroller and triggers an interrupt. The microcontroller inverts the signal and recheck (let's say after 100usec) the pin attached to the comparator. If the pin is low the short circuit is gone, if the short circuit persists shuts down the power to the motor drive.
Note the output of the motor drives is a digital power signal decoded by the train decoder, it isn't pwm.
Can this work? Can be easily improoved? I appreciate any comment.

parts:
psu
Buck converter
motor drive

 

Not sure I understand this or the motor control system - those controllers drive power into a motor, I don't get the train decoder bit... much more info needed.... but one thing I can say is that those buck converters have no reverse output polarity protection; Its unlikely the current limit will protect them if shorted together - the energy in the inductors is significant. But the way I see it the bigger problem is the motor controllers, its guaranteed somewhere along the line the high side of one will be on while the low side of the other is on - on the same side of the motor. Current-limiting there may not help either. Your microcontroller won't help, its far too slow, by the time its recognised a short the MOSFETs in one or both controllers could be toast.