It is not unusual to see AVR microcontollers in Boost / Buck converter designs.
This is form the ATMega328P data sheet.
The output voltage would be measured at an analog pin via voltage divider network.
The difference between the target voltage and the actual voltage would be calculated and an algorithm applied to change the duty cycle.
This is a teaching/learning project and may not necessarily result in the most cost effective solution and the student can't simply buy a buck converter from Aliexpress and submit that as his project work.
Actually I think it might be quite unusual. I'm no expert, but I've never come across a microcontroller in a DC-DC converter. It would seem a risky design choice: microcontrollers are at risk of crashing in a noisy environment; and software is at risk of containing bugs. Bearing in mind that voltage regulation, current limiting, under- and over-voltage protection, and over-temperature protection are really important in terms of protecting the devices powered by the converter, it seems crazy - in fact, bad engineering - to trust these to software when purposely designed and comprehensively tried and tested chips are readily available at very low cost.
Obviously it is possible to use an Arduino in a DC-DC converter. But it would be ridiculous to do so. Indeed, if one were used, there should be an exceptionally good justification for it.
The main point is that a microcontroller is simply too slow to successfully control a power converter by feedback. Switchmode power converters actually regulate on each PWM cycle.
But the consequences of a program crash, even in the absence of bugs, most certainly preclude using a microcontroller.
When the ATMega328P data sheet refers to "power regulation", it clearly means variable power control rather than voltage regulation.
At the risk of giving the impression that I am interested in continuing this part of the discussion, I'll just finish by saying this.
Microchip have published a paper on the subject of designing a buck converter using an MCU:
Anyway, let's not lose sight of the fact that purpose of an academic exercise is to expose the student to various technologies so he/she can gain an understanding of how these can be applied and the risks, benefits etc. of various solutions. In this case, the objective is clearly not to produce a marketable, cost effective buck converter.
