InLuvWJ's Blog

I'm glad I took the class overall… though certain things about it would have been better during a regular semester: only 5 weeks was hard to get things done in and with taking another class (4 credits) it was hard to time manage. Over a regular time span it would probably have been easier to do. But it was still good. I saw a lot from other people's projects that I hadn't been exposed to before. Although a lot of engineering is still over my head, I learned a lot from listening to everyone present their projects. I like that it was a smaller class (not 40 people) so it was more personal. I got to know people I hadn't before. And although we didn't accomplish everything we wanted to with our hall sensor project, I'm proud of the work we did get done. I'm especially proud of Joe and Jay, as their work on the hardware took the most time and effort.

With our small scale model working, we only need to make the connection between the results and graphing those results. We're working on simultaneously stealing the IEEE lounge group's code for getting our outputs to the website and trying to make a bigger scale model. The sensor doesn't work from more than 2 inches away, and we've run into coding problems as well. Lately we found that our sensor doesn't debounce on its own. In testing the in and out variables counted up to 22000 people with a single finger swipe through the sensor. That is obviously a little off, so today we worked on integrating either a wait function or our debouncing code (in assembly) from our 222 days. Since Chris's graphing code works perfectly, we only need to send our results from our microcontroller to the site. Still in progess…

We had a group meeting on Saturday, which was good since it's so hard for us to all get together at the same time, what with class schedules and such.

Jay and Joe succeeded in making a dollhouse-sized version of what we want to put in the dining halls. When we walk our fingers through the sensors, the lights turn on. Sweet right? So the hardware is great, the code (Rob's code) is there, and I'm working on editing it to include:

• if statements (or switch) about more than one person crossing the sensor at the same time
• 1-2-3 vs 3-2-1 tripping order (entering vs exiting)

Chris and I are collaborating on getting the data collected from the hardware to the graphing code. This middleman is the biggest part of the project we have left to conquer.

The second week didn't get us very far. It actually presented more problems than we thought we had. How are we going to store all the data we get from our sensor? We're pointed in the direction of mysql…which I had never even heard of until it was brought up in class. Since the guys are handling the hardware right now, I'm looking into mysql, but I have no idea what I'm doing with it at this point. Lots to learn in this class! So once we have a trip sensor, we need the code to recognize if it was a person entering or exiting, and count how many people are now in the room. AND we need to be able to store that information so that we can find it 3 weeks from now if need be. Learning learning learning…

The first week is only the planning stage, and good thing too. Our original plan to make an automatic door locker/unlocker for a house door didn't make the cut because apparantly that already exists. Huh, who knew? So now the five of us are piggybacking off the tripwire project from last winter. We need to obtain all the parts (a list of which is drawn up), and building will hopefully commence on Monday. This was decided at a group meeting on Saturday (which I phoned into). For now the goal is to learn the code backwards and forwards and add onto the hardware schematics from last year.

You can see what we've been working on at our Project page