Yurtle
Robot built for 'Into the Deep' 24-25 competition for FIRST Tech Challenge. Reached playoffs at Chesapeake Championship and selected for national robotics premier event.
Yurtle is FTC 24158 Green Gang’s robot for the 2024–25 Into the Deep season. As the software lead, I built the full control stack from the ground up.
The centerpiece of our software is a Finite State Machine inspired by the Roadrunner library, used for both teleop and autonomous. In teleop, the FSM encapsulates robot state to simplify and automate subsystem transitions, scoring, transferring, and intaking all trigger automatically from a single button press. In autonomous, FSM-based state tracking lets us repeat movements and build in failsafes if any path segment goes wrong.
To handle a 5-subsystem transfer sequence, I implemented a Java Commands framework that chains subsystem methods simultaneously or sequentially, abstracting all individual movements behind one driver input. I also wrote a custom GreenLinearOpMode base class to standardize initialization, reduce code duplication across opmodes, and expose configuration toggles (field-centric vs. robot-centric drive, alliance color) that integrate directly with our intake’s color sensor to auto-eject the wrong-colored samples.
Subsystem precision is handled through PID control on the intake arm and motion profiling on high-DOF subsystems, preventing robot tipping at full vertical extension and enabling smoother, faster cycles. For localization, we use Pinpoint odometry, switched from two-pod dead wheel tracking before Michiana for its faster refresh rate and more accurate IMU fusion. Autonomous paths are built with Road Runner spline trajectories for continuous, smooth movement with full control over position and heading throughout each segment.