What does the robot need to be able to do?
This is the first step of our engineering process. As a group, we are figuring out exactly what our goal is, both immediate and final.
Our goals are what we need to do and time it might take. We are designing to specs. We want to be able to measure when we meet our goals.
This will be helpful in the future when comparing which prototype is better.
Room, with SMARTBOARD and Whiteboard with ideas on them.
Mechanism brainstorm
We needed to get our ideas on paper. We came up with an idea to build the general chassis and prototype different ways of movement. This is completely new for our team.
Whiteboard with Mechanism Ideas.
Tank tread prototyping
Shanti and Wilson worked on building preliminary tank treads to test. Currently, Tetrix is a pain. The current treads will be very basic. They are in the middle of working on it right now.
They are using Tetrix, plastic sprockets and plastic tank treads to test. There wasn’t much time to test, but the one side that had been constructed looked to work fairly well. There is no traction right now.
Next meeting, the other side has to be built and the treads need to be tested more.
Whiteboard with Mechanism Ideas.
Image processing
Matt and AJ worked on the particular elements of image processing that will be needed to identify the beacon in autonomous. In particular, they started work on a program that will be able identify a blue region, isolate all the image processing work to the relevant region of the bitmap, and then identify the black button. While OpenCV is a powerful framework, it seems like it may be overkill for this particular problem. It also suffers from poor documentation, along with the fact that the OpenCV API was obviously ported straight from C, and so retains some idiosyncrasies from C that a native Java API would not force the programmer to deal with.
Coders Working on Image Processing.
Prototyping a fan
Isaac did a test by blowing air on the balls and they rolled away well. So, he figured that it could be scaled up and a fan could do the same thing. He began by designing a good gear ratio to spin the fan as fast as possible. The ratio is 9:1. By accident, he geared at backwards so it actually went 9 times slower, but that was easily fixed. He constructed it by heating some ABS and bending it into a propeller, using Tetrix for speed. He tested on the field and he was able to blow some of the balls, but one of the balls surrounded by a couple cubes would not move.
Whether this was the way the ball was on the floor or because it was surrounded needs to be tested in the future. He is building another iteration with four times more speed that will hopefully be able to blow balls even when they’re in different orientations.
Isaac Making a Fan.