denkbots’ cRi3D Reveal
We used the tools discussed in Part 1 – Game Theory to determine our strategy in Part 2 – Strategy and Research which allowed us to defined our Robot Requirements in Part 3 – Robot Requirements for this season’s game (and game manual). Now we are ready to build a robot.
To keep this conceptual Robot in Three Days as approachable as possible, we are going to start with the AM14U3, 2016 KoP Chassis (am-14u3) from AndyMark. In order to fit under the Low Bar, but maximize our ability to complete other DEFENSES as needed without any compromise to our initial Robot Requirements, we are going to retrofit the chassis to run 4x Pneumatic Tire 12-1/2 x 2-1/4″ (62-203mm) from AllegroMedical. This gives us more than enough ground clearance to traverse all Category B (Moat and Ramparts) and Category D (Rock Wall and Rough Terrain) DEFENSES while still allowing us to fit under the Low Bar.
Next we will setup our Boulder intake. To ensure the most efficient cycle time, we will design our Boulder intake to collect balls in the “back” of our robot and shoot them out of the “front” of our robot. This requires extra driver skill, but designing out unnecessary turns from our cycle will increase our cycle efficiency. By removing the back bar of our chassis and supporting it further in, we can create an intake space while maintaining the structural integrity of our chassis. Using a simple horizontal roller (here is an example found while researching existing technologies – FRC Team 100 Rebound Rumble 3-Ball Intake Test) we will take in the Boulder and hold it in the back of our robot until we traverse the Low Bar and are ready to shoot.
Finally, we will shoot the Boulder into the High Tower Goal. By rolling the Boulder up and out of our intake system and letting it fall down into a single powered roller/backstop shooter (here is an example found while researching existing technologies – FRC 3464 Shooter and Lifter- FRC 2012 Rebound Rumble) once our vision system has confirmed we are at a proper distance and angle.
This robot can DAMAGE three DEFENSES in a match, is optimized to have a cycle time of 10 seconds (13 cycles), and can park on the BATTER at the end of a match. Furthermore, because of the optimized intake/shooter system and vision system, it is possible that we could achieve a very special feat in Autonomous Mode. If we started with one Boulder in the Neutral Zone facing the Low Bar, drove forward and shot the Boulder then drove in reverse to the known location of a Boulder, brought it into the robot with the intake system, then drove back forward and shot; A DOUBLE BOULDER AUTONOMOUS ACHIEVEMENT could be unlocked!
Applying our game theory tool to this conceptual robot, we could be able to achieve the following:
- “cRi3D_MAX” Autonomous Mode: 30 pts
- “cRi3D_MAX” Teleoperation Mode: 90 pts
- “cRi3D_MAX” End Game: 5 pts
The “cRi3D” could be capable of achieving 125 pts by itself in a match.
If we take a conservative estimate, with only a single Boulder Autonomous Mode and a 15 second cycle time (8 cycles), we can achieve the following:
- “cRi3D_MID” Autonomous Mode: 20 pts
- “cRi3D_MID” Teleoperation Mode: 65 pts
- “cRi3D_MID” End Game: 5 pts
The “cRi3D” should be capable of achieving 91 pts by itself in a match.
If we evaluate the Worst Case Ontario, with a broken shooter we can still dump a Boulder into the Low Tower Goal and safely assume a 15 second cycle time (8 cycles), we can manage the following:
- “cRi3D_MIN” Autonomous Mode: 15 pts
- “cRi3D_MIN” Teleoperation Mode: 41 pts
- “cRi3D_MIN” End Game: 5 pts
The “cRi3D” should be capable of managing 61 pts by itself in a match.
Hopefully these articles can help new teams, teams in transition, or teams that are just looking for a new and improved way to approach robot design to learn and apply an analytical requirements based approach this season. If you have any questions, feel free to reach out to us on Facebook or e-mail us at denkbots(at)gmail(dot)com!
As always, feel free to join the conversation on our Facebook or Twitter with your questions, thoughts, and feedback on these articles!
*Updated 20160113 to fix REACH vs CROSS scoring error