WHAT A DAY
I got up at noon and gave myself an hour before starting my project again. At 1pm I began using Matlab to enter all the rules and membership functions I had. It’s actually quite a nifty little near program. I can see why it’s a popular analysis/modeling/simulation tool. It’s unfortunate I don’t have any time to play around with it more right now.
Anyways, it took forever to get all the rules and membership functions in. I took a lot of care in not fucking up so that a little error would later cause me grief during simulation. I also had to eat lunch and I slacked near the end when I thought things might actually go ok. It was around 5pm when I started Simulink to build my simulation model. It took me another hour to figure out how the hell to draw models properly in Simulink. I was totally stumped at one point. In desperation, I called robotics lab and luckily Dana was there. She answered my question in two seconds and literally saved me hours of time. No wonder she’s going for her Ph.D and I’m dicking around with my Master’s.
Anyways, with my model and controller done I was ready to simulate everything. This represented a big monumental moment in the project. It’s like when you flip the switch for the first time on this complicated machine you’ve built. Usually, one of two things happen at this point: nothing happens or it totally blows up on you. When I pressed the Start button, it blew up on me.
The Matlab command window lit up like a Christmas tree. I got warnings and errors all over. I felt despair. It was around 6pm at this point. I didn’t know what to do and I was getting hungry. I decided I was going to get some food and come back and start again.
Half an hour later I was back at it. From 7pm to around 11pm was probably one of the most painful 4 hour blocks I’ve had all year. Nothing was working. At 10pm, I nearly reached a full-scale panic. I was busy calculating how much I needed to get on this project to just pass the course. I was convinced that I was not going to get this thing to run. I began to wonder if people ever survived failing a course in grad school.
Then, sometime around 11pm, I pessimistically tried to run the sim again, when something amazing happened. It didn’t spit out errors the second I pressed the button. The damn thing kept on running. I was afraid an error was going to pop up before the intended sim end was reached. It was like watching the end of a horse race. “Go you sonovabitch! Go!”, I yelled.
It made it. I clicked on the response curves. They looked reasonable though there were some pecularities near the end. I didn’t care. It was good enough. I began to whoop and holler in my room.
I’ve spent the last five hours fixing things up and writing up the results in my report. It still doesn’t run that great, but I don’t give a rat’s ass.
I’d like to think I’m done for the most part, but that would be erroneous. My prof wants us to also implement a conventional controller and make comparisons. I have no fucking clue how the hell I’m going to do that.
So, when I wake up tomorrow, I have to accomplish what I did today, but only this time in a different way. Excellent.
Honestly, this is all fucking stupid. Other people are having fun tonight, drinking, partying, watch Punky Brewster re-runs, having sex, and I’m running my crapping controller in Matlab. It’d be different if I was curing cancer or making ears grow on monkeys, but is just a project that will benefit no one. I’m not expressing my displeasure with school in general, but I’m saying there’s got to be a better way to learn or assess my learning other than making me a prisoner in my room 72 hours. No one adequately learns intelligent control in three lousy days.
I tell ya, come 4pm Monday, I’m going to be one happy bastard.