Make or Break Week

This is my make or break week. I need to polish up the work I’ve done so far this summer, update my thesis proposal document, hopefully not have to update the original one with the department, schedule my thesis proposal before the end of the month, and make sure I complete 5 days of running one hour a day. I also need to figure out what’s wrong with my landlord’s car. I’ve had it for an entire month. It’s time to either sh*t or get off the toilet.

School is tricky for me now more than ever. There are days when I sit down and power through a lot of work. Followed by plenty more days where I hardly do anything. I’ve definitely lost whatever mojo used to get me through this. Either that or I really screwed up with a project that in the end doesn’t capture my interest as much as it should. With my Masters, I was just as much on my own (though now that is normal) and while it was tough, what really got me was applying high level concepts to solving the same problems I’d have to deal with when building robots on my own.

For me, it’s like this. I’ve always been interested in balancing robots. For some reason, the mobile inverted pendulum has really captured some nugget of my soul. My first attempt at building one utilized distance sensors to measure the distance to the ground. Pretty much, my thought process was that if I kept the sensor at the “balanced” distance at all times, then it would work. It was a very simple control system, and after fine-tuning, worked well, so long as the floor was level. But that was about it, the robot would move around, balanced, but not controlled in any other way and would eventually crash into things.

The next iteration used two distance sensors. This helped cancel out bias towards the direction where the single sensor was located, especially when not on a level surface. Allowed me to come up with some smarter methods to control movement, a bit more stable. Then I learned of PID controllers. Now, not only did my controller respond to the error, but it also would account for the derivative in the error (how fast the error was changing) and accumulated error. I could go back to a single sensor setup yet have significantly better control over balance and movement. And when I started including modeling – the physics of the inverted pendulum, translating the distance measured to tilt angle and incorporating that information into the design, then I was really cooking with butter!

That’s how I felt with my Master’s work. I’d taken high level concepts used to solve other problems and started figuring out ways to get faster and much better solutions to problems that I could already solve. And while my existing solution may seem good to me, especially given what else was out there and the effort I put it in to the reward I got, it blew my mind that by just studying and learning a bit more, I could come up with something orders of magnitude better. And so despite the frustrations and times when information seemed to refuse to sink in or be totally over my ken, I’d stick with it and plow through, working, trying out things every day.

But I just don’t have it now. What excites me are smaller problems. Subparts of the whole. There is something I need to do or accomplish – for example and equation to solve, but for whatever reason, my setup isn’t in a standard form or there needs to be some tweak so that I can solve the problem using known methods – that excites me. But the overall project just doesn’t drive me as it should. But I’m so close to being done that it is too late to start over.

On another note, I’m completed two days of the running. I weighed in this morning at 206lbs. Still not my lowest since I started the diet, but I’ll take decrease from my NYC trip. And speaking of the diet. . . well, the ultra low carb part seems to be over. While I’ll try and more or less keep my carbs low, which will be easier once I do my beginning of the month grocery shopping, I’m hoping that with the increase in running, I don’t have to stay as strict. That said, we’ll see just how well this works out.

Speaking of the running, I still only manage to hit just over 5 miles in an hour. My 5k time is around 34 minutes, depending on whether I did any weight training before I started running (drops to 36 minutes, best has been 33). I think my body may be more attuned towards high intensity interval training. So perhaps Wednesday, I’ll give that a try by pushing my “max” speed up some more and alternating strictly one minute max, 2 or 3 minutes min (fast walking) speed and see if that improves my time and distance.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: