On this blog, members of the Midnight Mechanics FIRST Robotics team will daily update on what it takes to be a FIRST team.
DRIVE TRAIN-
Progress today! We have the pan and tilt servos working independently, but not together. We also programmed this year's robot, but had problems because we want two RCs to work with; the one on the 2006 robot and the one on our testing 2005 robot. However, there were problems in loading code into last year's microcontroller. We also moved code into a different function to be called from autonomous mode. We now have two working dongles, thankfully. Testing has been a lot easier, and we're getting closer to optimal gain constants and such. It's getting really close to ship date, and we're more behind than we want to. At least we're driving!
Today, we worked out the bugs in yesterday's programming. We figured out that the floating point variable was being put back into an integer variable, which made the number wrap around and not work. We had battery problems after that. We also put together the dongle, which still needs repairs for the autonomous mode, but the disable button is working well. It has been quite useful for trying out code on last year's robot. Tomorrow, it should follow the green light. We need to write the code for drive mode, and that's what one of us will probably be doing tomorrow while the rest work on the camera code. If time permits, we will work on the shooting mechanism, but it probably won't be in the autonomous mode because it takes too long.
In the first half of the meeting (before the 5:30pm bus), we recapped the last two meetings and split into three one-person teams for controller mapping (deciding which buttons to use and how); sine, cosine, and tangent (figuring out how they work with our code and such); and dongle building (making something to simulate the game modes, like autonomous and disable).
Today, we worked on getting the camera to move the robot accurately. We wrote and compiled the code today! Another milestone, but not as huge. Although it's going in somewhat the right direction, we finally figured that have problems with the trignometric functions at the end. We fixed most of the other problems as far as syntax and random other problems go. Once we get this fixed, onto the shooter we go!
Today, we worked on coding the joysticks as inputs to the Jacobian Controller (the algorithm that translates the x and y desired locations to a seamless robot motor movement). That now works. We also worked on code to translate pan and tilt servos to the inputs for the Jacobian controller. We have not tested out that code yet, nor have posted onto svn, but it is theoretically supposed to work. Next time we meet (Monday), we will test out that code, and modified as needed. The next goal is to get sensors onto the shooting mechanism to determine the shooting angle.
Today Kyle and I worked with ball gathering. We helped mounted the motor onto the ball gatherer. We soon ran into the problem that the sprocket was not aligned with the sprocket that was on the ball gatherer itself. So we drilled new holes so it would line up, but we ended up drilling holes that we a little off. We ended up cutting a new plate for the motor. Kyle used the jigsaw to cut the plate and I drilled in new holes into the plate. Briseida, Kelly, and Guillermo worked on threads for the wheels.
We are now working on code to translate the pan and tilt servo angles to the distance needed to move the robot. It's a work in progress, although now we have code for distance to motor calculations done. After finishing this code up, we need to work on how to shoot the ball. We still need to see if we can use sensors on the shooter and how to compute this. We only have about a week left, so it's crunch time!
