CattyKid's Blog

I'm in the process of finalizing everything on our project page, and when that is done I will be helping to actually set everything up in the lounge. A lot has gotten done in the past few days.

After getting the IR sensors to work (in the lab, more on that in a bit) I turned my attention to a few other things. I needed to get the lights turning on and off in the context of the code, which took a while longer than I thought. There were some… ghosts in the code (mainly me missing a few lines and having too many files open at once) that were keeping the light from turning off. It also took a bit of fumbling around with the Vibration Sensor Sensitivity to get it to only trigger when it should.

The presentation last night in class went over really well. Our team worked together and shared the burden/time pretty well and I was proud of everyone. Luckily, that presentation was similar to what we will be giving tomorrow (except for the obvious absence of pictures, but that won't be too difficult to fix) so we have a great foundation for that presentation already.

We ended our presentation last night by sending blackthroat down to the IEEE lounge and actually playing around with the microcontroller in order to show the website in class. On the advice of afterburn, I added a little bit of color to make the page more readable and a little more pleasing than all black and white.

Now, about the IR Trip wires. It seems that what we have thus far is pretty accurate, and time is actually quite tight. I'm not sure if we are going to be totally integrating them into the system for the final presentation. All of the components are working, but we are using command hooks and tape to hold everything up in the lounge. Adding the IR sensors, I think, is going to make everything look a lot sloppier than it needs to and will only offer minimal benefit. If someone moves enough to set off an IR Trip, they have also moved enough to set off the PIR sensor. So in short, I don't think they are actually going to help our project at all, so I don't think we should put them up.

It is important to note, though, that the IR Trips would have some use in slightly different applications. For example, we are using a reed switch on a door. There are some rooms in some places (think your living room or your kitchen at your house) that do not have door. Obviously, a reed switch there is going to be completely useless. A much better solution would be to use an IR Trip Wire because it is unobtrusive and cheap.

So, a long blog entry but that's fine since it contained some important stuff.

Off to finalize the room and presentation.

So it seems like tonight my blog was much more of a Twitter feed than anything, but a lot of important stuff got done tonight, so that's great.

So, the last thing I did was put get the IR Trips to work with the existing code. It was actually really, incredibly easy. All it involved doing was copying and pasting my already working code form the other program and, wherever the instance TPM1 occurred, I changed it to TPM2, making sure to pay attention to caps.

So, tomorrow, here we come.

More good news. So, I was having some trouble with getting pySerial to give me up to date statuses on successive serial port reads. For some reason, I was getting the same value that was read when the program was first run. No amount of successive reads would give me an up to date status, even though it was outputting correctly (as confirmed by minicom and hyperterm).

So after doing some research I found a few important things: * How to get a USB to serial converter working under Ubuntu : http://forums.reprap.org/read.php?12,4546

• Got pySerial to read things from a USB port too (after the serial adapter was initialized, of course) http://www.mwgaa.com/blog/2010/01/17/gumstix-eat-your-pyserial/

So, now I am working on putting some more of the stuff together and moving closer to project completion.

Well, hallsensor team was having a little bit of trouble. Since we needed these sensors too, I put some time into it tonight. I got them to work pretty well at about 2 meters, using their TPM code and my hardware. So, that is great news because now they can progress and we can get started finishing everything up.

The detectors are working at a distance of about 2 meters. We are not getting the best binary (1 or 0) output, more of a bouncing input between 1 and 0. But for us, that is actually pretty okay. If we get any 1 when we sample, we know someone is in the room. So, the bouncing isn't too bad for our application. Additionally, the retroreflectors seem to be nearly useless and refuse to reflect anything.But that is a compromise I am willing to make at this point.

So I am having some problems with python and my program that are really confusing me. The microcontroller outputs a string of the form

”/ 1 1 1 0 0 0\n\r” where the first ” signifies the beginning of the string and the second ” signifies the null character, aka the end of the string.

So, these are being stored in the array line, and in python array indices begin at 0. So, in order to access individual numbers (actually characters in a char array) I should type: line[2] line[4] etc, but for some reason i need to access them with line[3] line[5] etc. Other ways will not work. So that is nothing short of very weird.

Then, we are using a program called pyserial in order to read from the serial port. Successive reads from the port are always the same as the first read. Even though hooking the microcontroller up via hyperterm or via minicom shows correct strings being sent from the controller, my python program will not see anything besides what is sent the first time. Going to need to talk to afterburn asap tomorrow to get some things worked out.

Other than that, I am actually feeling pretty good about the project. I mean, we still have one or two more things to do, but I am not too worried about them. Tomorrow, the amplifiers will be coming in, so we can get the vibration sensors working a little bit better. They should not take too long to build. Also have to get the IR trips working, still waiting on the other group to get that hammered down. If they don't have it by tomorrow, maybe our group can take a look at it from a new point of view and come up with some way to help. A fresh view never hurts. Perhaps they are using a voltage or amperage that is too low? Maybe we might need to use more than one IR LED firing to get some more light?

So, it is about time to start getting the presentation together.