Our robot for 2014 was Katrina, named after the large catapult that made up most of her game piece. Katrina (or Kat)'s task was to intake a ball from a human player, carry the ball a short way down the field, before throwing it either over the truss or into the goal. Aerial Assist was an extremely dynamic game, and Kat was designed to reflect that, having the ability to catch a ball thrown by a human player, pick up the ball from a ground, pass to another robot, throw across the truss, and score a high goal (in both teleoperated and autonomous mode).
The game for the 2014 FRC competition was Aerial Assist. The aim of this competition was to build a robot that could move a ball from one end of the field to the other then shoot it into the goal. The catch was, that most of the points came from cooperative with your alliance members with points being awarded for controlled passes between robots across the field. This meant that the robot had to be maneuverable and be able to intake, pass, throw and preferably catch the ball to perform as an offensive robot.
Katrina’s chassis makes use of the provided kit drive base with the addition of an off the shelf mecanum upgrade kit. This facilitated omnidirectional movement (moving in any direction), chosen to to give the driver maximum flexibility while on the field. It proved not to be terribly effective as we had to compromise on speed and the mecanum system overall was not very efficient. This drive base arrangement was something the team had never done before and proved to be a very worthwhile learning experience for both mechanical and software. The system presented an interesting controls challenge that we met with the use of gyroscopes and accelerometers to improve the driving accuracy.
Kat had to be able to pick up the ball off the ground and then fire it into the high goal accurately and repeatedly. To throw the ball we used a catapult, the power came from six trampoline springs anchored to the front of the robot. The catapult arm was drawn using a worm drive winch powered by a mini cim at 44:1 gear ratio. The arm was held in place by a quick release so that the witch could be un-wound and the catapult fired. To pick up the ball we drew inspiration from one of the Vex robot in three days teams and decided to build chopsticks. This are two counter rotating sticks that protrude from the robot frame, then can be raised and lowered are and are pulled towards each other by springs. This method was chosen because the ball was wider then then our robot such that we could not have a rigid frame that went around that ball as this would have been outside our frame perimeter. The chopsticks proved to be no end of trouble and ultimately a poor design decision. However we were able to pickup the ball on a few occasions and they did allow us to catch the ball from the human player. They will always be remembered as the teams biggest lesson in good design and critical analysis.
When developing our robot, we made one critical error. We aimed to create a robot that did everything, not a robot that did a single task reliably. When it came to competition, our initial strategy was to act as an offensive robot, receiving the ball from allies then firing it into the goal.
In reality we had extreme issues collecting dropped balls from the ground, and so our strategy was swiftly changed so that we acted as an intaking robot, receiving the ball from the human player and passing it onto our allies. We worked effectively in this position, but not as well as robots who were designed to operate here.
During the Autonomous period, our robot would drive forward and fire a ball into the goal without regard for whether it was “hot” or not. During the Hawaii regional our accuracy was less than 50%, however after software modification, our robot achieved 100% accuracy at the Duel Down Under.