In competitive robotics, the autonomous period relies heavily on high precision movement and minimal drift. The challenge was to create a flawless "red auto" routine capable of navigating specific field markers efficiently, accounting for mechanical slip and electrical latency.
The primary hurdle was mechanical unreliability during high speed strafing. To resolve this, I advocated for and integrated goBilda Rhino wheels, dramatically improving the friction coefficients of our drivetrain. On the software side, standard time based logic was insufficient. I transitioned the team to coordinate based movement mapping using the Roadrunner library.
By relying on dead wheel odometry rather than raw encoder ticks, the routine achieved a near perfect consistency rate. The integration of advanced mapping dramatically increased our autonomous point yield.
Next steps involve utilizing machine learning for active game piece detection, replacing static coordinate arrays with dynamic pathfinding calculations.