AlphaBit - OpenML

Autonomous Control - Getting Started

Autonomous Starter Guide

This page is designed for FTC teams that are starting autonomous implementation. Follow these steps in order to build a stable baseline before advanced optimization.

Quick Intro

Odometry Pods Before Routes

Localization quality decides whether your autonomous will feel "random" or repeatable. Most teams choose one of these layouts:

Two pod odometry FTC diagram — 2-pod + IMU: simpler, lighter, easier to maintain.

Three pod odometry FTC diagram — 3-pod: stronger pure odometry heading estimate, more mechanical complexity.

If your team is beginner-level, start with 2-pod + IMU and move to 3-pod only if needed.

Step 1

Prepare the Baseline Autonomous Class

Main reference file: drive/Autonomous/AutonomousControl.java (regional code).

Keep one minimal route first (score once, park safely).

Do not start with 4 full routes and cycle logic on day one.

Step 2

Implement Case Selection and Start Poses

Use a controlled init loop and switch statement to choose route/case:

while(opModeInInit()) {
    if (gamepad1.dpad_left) { /* toggle alliance */ }
    if (gamepad1.dpad_up)   { /* toggle side */ }
}

switch(autoCase){
    case 0: drive.setPoseEstimate(startPose_RedAudience);
            drive.followTrajectorySequenceAsync(trajectoryRedAudience); break;
    case 1: drive.setPoseEstimate(startPose_BlueAudience);
            drive.followTrajectorySequenceAsync(trajectoryBlueAudience); break;
}

Step 3

Attach Mechanism Actions With Temporal Markers

Use trajectory markers for intake/turret/shooter timing.

Avoid long fixed sleeps for mechanism orchestration.

Keep failsafe flags for timeout recovery.

Step 4

Add AprilTag Pattern Support (Optional Early)

Pattern IDs 21/22/23 can refine audience-side logic.

If detection is unstable, run a fixed fallback route.

Step 5

Run Validation in Three Layers

Layer A: no game pieces, just path endpoint accuracy.

Layer B: add one pickup + score cycle.

Layer C: full cycle and match-like timing pressure.

Common Fails

Odometry calibrated after route tuning (wrong order).

No fallback case when vision fails in init.

Too many moving subsystem changes per test run.

Open Odometry Pods for the detailed calibration workflow and 2-pod vs 3-pod tradeoffs.

Support -> Discord

Choose Language / Alege Limba

AlphaBit AI Assistant