This week was particularly crunchy for me in other classes, but I was able to make some headway coming up to solutions for problems we were going to have to deal with later down the line.
One of these was the problem of led positions. We’re dealing with an organic layout for our LEDs, as opposed to the typical matrix that a lot of applications are accustomed to. Given this I developed some code that allowed for visual effects that depended on hard-coded positions rather than relying on matrix-style modulus operations.
I also put some thought into the problem of smooth transitional effects as prompted by Joe Rowley. Basically the problem is that given the capricious nature of user input, we need to develop a system that allows for smooth transitions from one behavior to another in a way that doesn’t break the user experience. Joe suggested multithreading, but on Arduino I don’t think that’s really an option, and implementing it from Raspberry Pi doesn’t feel clean. Instead I was thinking about hard coding a dispatch system into the Arduino that would keep track of the states of different effects and pass the buck off to whatever is needed. The math shouldn’t be that difficult, we just need to average out the two states according to some weight that would move at a speed proportional to its urgency. The problem is that I’m not sure if doing this would be fundamentally different in Arduino vs. Processing, so I have no prototype and the idea is just in my head for the time being.g.
I’m really glad that Gene got us a working LED board prototype to test stuff on but I’ve just been to busy to get into the lab this week, with that on top of the fact that Rowley is the only one who really understands how to get the LEDs going. I’m in the process of setting up my new raspberry pi. Next step is to see if we can get the web-pulldown scripts working on it and see if we can forward it off to the arduino. My goal is to have some visual representation of the oceanographic data displayed on the Arduino by next week.