Limelight 3A FTC Training

1. What Is the Limelight 3A?

The Limelight 3A is a USB vision camera with its own processor and browser-based configuration interface. It can run AprilTag, color, barcode, and neural-network pipelines. It does not have built-in LED illumination; check the camera image and exposure settings for your lighting conditions.

• AprilTags
• Game elements
• Targets or markers
• Objects you train it to recognize

FTC teams can use its target measurements for alignment or use its AprilTag localization output as one input to autonomous navigation. Camera results are measurements, not a complete robot-control system.

2. Why Teams Use Limelight

• Runs vision pipelines on the camera instead of doing all image processing on the Control Hub.
• Can report target offsets and AprilTag detections.
• Can estimate field pose when the correct AprilTag map and camera calibration are configured.
• Still requires testing for lighting, mounting, latency, and field conditions.

3. How Limelight Connects to Your Robot

For FTC, connect the Limelight 3A to the Control Hub with a USB-C-to-USB-A cable in the blue USB 3.0 port.

• Connect it to a computer to configure it at http://limelight.local:5801 or through Limelight Hardware Manager.
• Set the team number, configure a pipeline, and restart the vision client.
• In the Driver Station Robot Configuration, scan for the device. It appears as an Ethernet Device; rename it limelight.
• Use that configured name in hardwareMap. The 3A uses USB-C and has no RJ45 port.

4. What the Limelight Sends to Your Robot

The Limelight gives your robot important information such as:

• tx – horizontal offset (left/right)
• ty – vertical offset (up/down)
• tx and ty are horizontal and vertical target offsets in degrees.
• ta is target area as a percentage of the image.
• Fiducial results include AprilTag IDs and per-tag measurements.
• Robot field pose is available only when the AprilTag pipeline, field map, and camera pose are configured.

Your robot can use this to turn, drive, or align automatically.

5. Read Results in FTC Java

This LinearOpMode example polls the camera and reports valid target measurements. Set the pipeline index to the pipeline configured in the Limelight web interface.

import com.qualcomm.hardware.limelightvision.LLResult;
import com.qualcomm.hardware.limelightvision.Limelight3A;
import com.qualcomm.robotcore.eventloop.opmode.LinearOpMode;
import com.qualcomm.robotcore.eventloop.opmode.TeleOp;

@TeleOp(name = "Limelight Check")
public class LimelightCheck extends LinearOpMode {
    @Override
    public void runOpMode() throws InterruptedException {
        Limelight3A limelight = hardwareMap.get(Limelight3A.class, "limelight");
        limelight.setPollRateHz(100);
        limelight.pipelineSwitch(0);
        limelight.start();

        telemetry.addLine("Limelight ready");
        telemetry.update();
        waitForStart();

        while (opModeIsActive()) {
            LLResult result = limelight.getLatestResult();
            if (result != null && result.isValid()) {
                telemetry.addData("tx (degrees)", result.getTx());
                telemetry.addData("ty (degrees)", result.getTy());
                telemetry.addData("ta (percent)", result.getTa());
            } else {
                telemetry.addLine("No valid target");
            }
            telemetry.update();
        }

        limelight.stop();
    }
}

6. Using Limelight for AprilTags

The Limelight 3A can detect AprilTags automatically. This helps your robot:

• Know which side of the field it’s on
• Choose the correct autonomous path
• Score more accurately

AprilTag detections are returned as a list. Check the configured pipeline and use the actual tag IDs for your field/game, rather than assuming IDs 1, 2, and 3.

import com.qualcomm.hardware.limelightvision.LLResultTypes;
        import java.util.List;

        List<LLResultTypes.FiducialResult> tags = result.getFiducialResults();
        for (LLResultTypes.FiducialResult tag : tags) {
            telemetry.addData("AprilTag ID", tag.getFiducialId());
            telemetry.addData("Tag tx", tag.getTargetXDegrees());
            telemetry.addData("Tag ty", tag.getTargetYDegrees());
        }

7. Localization and Safe Testing

For field localization, enable Full 3D in the AprilTag pipeline and configure the camera's position relative to the robot. MegaTag 2 also requires current robot yaw to be sent to Limelight before reading its pose.

// Send the robot's current heading in degrees before reading MegaTag 2 pose.
limelight.updateRobotOrientation(robotYawDegrees);
LLResult result = limelight.getLatestResult();
if (result != null && result.isValid()) {
    Pose3D fieldPose = result.getBotpose_MT2();
    if (fieldPose != null) {
        telemetry.addData("Field pose", fieldPose);
    }
}

Do not combine this pose with your autonomous localizer until you have checked coordinate frames, camera mounting, heading conventions, and latency against known field positions.

8. Practice Challenges

• Use Limelight to center your robot on a target
• Use the detected fiducial ID to select between team-defined autonomous options
• Stop the robot safely whenever the result is missing or invalid
• Print Limelight values on telemetry

9. References

Limelight 3A Quick Start

Official FTC Java Programming Guide

FTC SDK SensorLimelight3A Sample