‘Bot in Action
Yesterday I added training wheels to the bot and allowed it to run free in the kitchen. The “training wheel” is a simple furniture caster attached to a 1.5″ wide piece of aluminum bent in two places and bolted to the frame. The controller is running a very simple navigation program. It basically goes forward for two seconds, turns right for one, repeat. The timing is too far off to make any kind of recognizable shape, but it was a chance to see the ’bot in action.
There’s also a short high-power burst forward and backward (and forward again) to qualitatively determine what kind of torque characteristics the system exhibits. Interestingly, it’s enough to kick the caster wheel up off the ground, despite having two battery packs strapped to it.
For its first untethered tests, I added two 6-cell, 3000 mAh NiMH battery packs (the kind used in electric R/C race cars and planes) wired in series to give a nominal battery voltage of 14 V. This makes for rather large drop across the controller’s main regulator (and might even be outside the spec; I’ll have to take a look). Perhaps I should tap off of one of the batteries for powering the electronics.
PID Controller
What a crummy day. I spent so much time fiddling with the LCD code that I broke. I had started writing a basic PD speed controller. I was writing a fair amount of info to the LCD during the PD loop interrupt, which was taking quite large portion of the available executiong time, so I tried to tighten up the LCD code (more details in a later post). Well, I broke it. Many, many hours later, and after fixing it and having it all of a sudden stop working multiple times, I got it and the PD code working.
Now, I based this off an old article written by David Anderson. It does something interesting. If you command a speed of zero, and then turn a wheel by hand, it will resist that effort like a spring. If you then release the wheel, it will rotate back to its approximate starting point.
The calculation of the PWM signal is cumulative, even if there is an error of zero, so the PWM is being integrated (I think). I need to brush up on my understanding of PID controllers to figure this one out.
MMC Works
Okay, I’m a dork. I made so many mistakes trying to figure out this serial communications thing, it’s not even funny. But, I’ve got it working now.
Two major things went wrong. First, I failed to realize that the ’32, by default, operates off of its internal 1.0 MHz oscillator. You have to program clock selection bits as a separate step to get it to use an external crystal (which I had connected, but didn’t realize wasn’t being used). Since all of my baud rate calculations were being done assuming a 4.0 MHz clock, they were way off (which explained the bizarre timings I was seeing on the TX pin).
