It's amazing how things work perfectly whilst in testing and then fail miserably in the field!

With no debouncing code the input just has to be valid for say 1microsecond for the cpu to say it's active - this time is in the realm of glitches and spikes so this may cause your motor to stop prematurely on random occasions thus making it hard to debug in the field. Also do you have current limiting on the motor drive circuit? What happens when you stall the motor? Does the cpu just keep trying to turn it in the direction waiting for the sensor to activate? Or does it have a timeout so that the cpu doesn't keep trying to power a seized motor and burn it out. When it does detect a timeout you can store an error code and alert the user. Does the unit 'fail safe'? When it does detect an error it should alert the user and do nothing more. What would happen to the vehicle if the dc motor kept on turning?

In the automotive circuits I've seen, things like ignition controllers sense if the controlling computer is alive, if not they have logic to bypass the computer control and use a fixed value for the ignition advance. In the fuel injection computers, if a non-critical sensor fails, the software flags the failure and substitutes a nominal value. If a critical sensor fails they go into limp home mode whereby some simple logic pulses the fuel injectors just so the motor keeps on going.

Ask the question 'what if' about everything your box does.
What if the motor seizes?
What if the battery voltage is too low?
What if the battery is removed and the alternator gives us 40v momentarily?
What if a sensor fails?
What if the ambient temperature rises above 60c?

There are a lot of conditions that can occur that you can't easily design around - in these instances your system must fail gracefully.

Sounds like a lot of work - better done now than when you find a problem in the field with 100's of pissed off customers! Also now would be a good time to review your liability if your design does fail. There is a difference between making a 'one off' for a friend and making a commercially viable product.