The author seems to be unaware that there are cheap stepper motor controllers available that handle all the tricky details of moving an axis to a given position smoothly. And they are implemented in hardware, so you don't have to sacrifice any CPU cycles to this task.
The bottom level tends to be basic Arduino shields that just switch power, but don't have a current limiting controller. Those need digital signal lines to tell them when to step. They waste a lot of power and can't be used with larger motors.
The next step up is a current limiting stepper motor driver.[1] These can handle more voltage and current, and have the sensing to avoid burning themselves out. You still have to provide pulses for each step. Go for at least this level. Stepper motors use considerable power when stopped, and you need current limiting on all but the tiniest motors. Otherwise you get to choose between overheating and too weak.
Next are drivers and controllers on the same board. These accept commands over USB, I2C, or a few other data paths, and take care of operating the motor.[2] If you're not into really low level programming, this makes life simpler.
I'm not really a woodworker but I have probably seen every Mathias Wandel video since his first marble machines. I wonder how much time I have spent over the last years watching his videos. It's gotten to a point where I know all technical jargon he uses like dovetail joints, tenon and mortises and whatnot. My brother is an actual woodworker but I can't talk to him about it, because I only know the English terminology and he knows the Dutch slang because our mother language is Dutch...
I want to enjoy his videos but something about his presentation style is very off putting to me. Friends often forward me one of his videos but I have given up trying.
You may still enjoy his Web page at woodgears.ca - that's mainly how I know him. Haven't seen more than a couple of videos, but I've extensively perused the site over the years.
I have visions of the author setting up a tracker with a simple ir motion sensor setup, testing everything works. Putting the heater on top, and then realising....
I'm kind of surprised these aren't more of a thing, for use in garages/cellars. Or for use in warehouses and other under heated work spaces.
I've been talking about this for fans for quite some time. Battery-powered fans could be smaller, and thus achieve better runtime on the same battery, if they'd point right at the overheated worker.
I've instrumented my place and used rrdtool to graph temperature in several places.
I found that the built-in forced-air heat (and a/c) seems to show up like a sawtooth wave, with an amplitude of about 3-4 degrees.
In the room with my computer desk, I've found that by turning off the forced air heat and using a vornado heater I end up with a very comfortable space with absolutely flat temperature graphs.
I use the one with the dial, not the LED controls, and it seems to have very good air circulation with well-tuned PID temperature controls.
One idea I had a while ago was that you could use these for saving energy by targeting exposed skin surfaces only. This would be particularly useful for offices: the office can be set to the lower temperature preferred by men, and then women can be targeted by the infrared heaters to keep them comfortable as well, so no one has to be uncomfortable while still saving energy. (It has to default to colder/male preferences because you save more energy by keeping the office as a whole at the lower temperature, and anyway, there's no such thing as an 'anti-heat beam' so you can't cool down a subset.)
Indeed, women in offices now tend to either suffer or sweater. They are solutions, but not good ones. Infrared trackers would allow one to have one's cake of attractive stylish (non-sweater) clothing while not suffering in the eating. So to speak.
If they were pervasive, they could allow institutional users to lower the temperature in their buildings a few degrees. One degree on the thermostat for a large building in New England could equate to a lot of savings in fuel.
It may surprise you but using a micro-controller and some servos is today often an easier and simpler solution than building a complex mechanism using cams and clockworks.
That sounds like a fun project, but economically impractical. When we installed panels on our roof it was cheaper to just lay them flat and buy extra panels to make up the difference, than to install them on static frames to angle them for most efficient energy generation. If static frames are not worth it, dynamic ones are definitely going to be too expensive.
Perhaps at scale, things would be different, but I doubt it.
Mathias's projects when he was still in high school during the 80's:
Primitive plotter: http://www.sentex.net/~mwandel/tech/plotter.html
Primitive 1-pin dot matrix printer: http://www.sentex.net/~mwandel/tech/printer.html
Primitive Commodore 64 drum scanner: http://www.sentex.net/~mwandel/tech/c64scanner.html
Home made wooden Joystik: http://www.sentex.net/~mwandel/tech/joystik.html