NXT Relay Robot

The following post is about a previous post I wrote about a NXT science project I worked on in 2013 for middle school science. There’s a long summary, and a video of the robot in action. (In the relay, it’s the first one, and that’s me setting it off.

Team GWEB (Graham, Woods, Ethan, Bennett) made it to the final four but didn’t go any farther. Our only success in the bracket was below.

 

If you watched the video above, I guess you get the objective. A robot relay. The first robot has to follow the line so it doesn’t hit the books (hit them and you get a 10, 5, or 1 second penalty, depending on how close they are to the starting line.) The robot has to reach the ramp (required to be 12 inches tall and have a 45 degree angle at the bottom) and hit down the pool ball. The ball must trigger the next robot, who has to avoid to books to cross the finish line.
We began with the basic line follower. How could you go wrong with that? I had worked way too hard on the last project to make a custom, completely from scratch robot. Doing this was honestly a pretty lazy approach, but  after my last project, I was glad to do this. My partner also had a custom bot on the last project, and was fine with building a robot that had instructions and a program already made for us.
We built the line follower, and it was surprisingly speedy. Problem is, when the robot loses the line, it’s set to search for the line in a clockwise circle. That’s fine for the first turn, a right turn, but the second turn is a left. The robot circles and circles but never gets back on track. That’s what we got for trying to take the easy route. We took apart the line follower and decided to build a custom robot.
Although Ethan and I were partners for our single robot, we had two other partners, Graham and Woods. We chose them as our partners that we would work with for the rest of the project. After we took apart the line follower, Graham and Ethan volunteered to go work on ramp, and Woods and I agreed to work on the robot together (Woods and Graham had the easier job, the second bot in the relay, so it only took them two days to build their robot.) It took two days to build the first gen GWEB (Graham, Woods, Ethan, Bennett), which was made only as wide as the NXT block (on purpose), and had a somewhat unstable but highly fast triple gear system. It took another day to make the gears stable, and another two to stabilize and balance the robot. From there on, we worked on programming.
We knew from the beginning that programming would be tough because of the speed of our robot – one motor rotation would send the robot a whole yard. We had to program down to the .001 rotation. Also, the battery level dramatically affected the robot’s performance. (We programmed using distance, no sensor.)  It took a week to get the robot consistent. After that, we didn’t touch it until the race, purposely leaving in the 60% full batteries. Our first round success is above.
Being the one seed from our 8.8 time trial, team GWEB got a bye after the first round, putting us right in the boys final four. Using the old batteries, our robot overshot on the turns by a mile. The inconsistency of our robot knocked us out.
Overall, I am happy with this project, even though we didn’t win. My group discovered how to create a triple gear system, and create a super fast robot. I don’t know what the next project is, but all that I learned in this project might come in handy for it. For my next project, it would be cool if I was paired with Graham, Woods, and Ethan again. Not only do we get along well, we are really good at collaborating.
Advertisements