Skip to main content

Robot Localization with MegaTag

If your Limelight's robot-space pose has been configured in the web ui, and a field map has been uploaded via the web ui, then the robot's location in field space will be available via the "botpose" networktables array (x,y,z in meters, roll, pitch, yaw in degrees).

Our implementation of botpose is called MegaTag. If more than one tag is in view, it is resilient to individual tag ambiguities and noise in the image. If all keypoints are coplanar, there is still some risk of ambiguity flipping.

  • Green Cylinder: Individual per-tag bot pose
  • Blue Cylinder: Old BotPose
  • White Cylinder: MegaTag Botpose

Note the obvious pose ambiguity here:

Notice how the new botpose (white cylinder) is extremely stable compared to the old botpose (blue cylinder). You can watch the tx and ty values as well.

This is not restricted to planar tags. It scales to any number of tags in full 3D and in any orientation. Floor tags and ceiling tags work perfectly.

Here’s a diagram demonstrating one aspect of how this works with a simple planar case. The results are actually better than what is depicted, as the MegaTag depicted has a significant error applied to three points instead of one point. As the 3D combined MegaTag increases in size and in keypoint count, its stability increases.

megatag botpose example:

Using WPILib's Pose Estimator​

info

In 2024, most of the WPILib Ecosystem transitioned to a single-origin coordinate system. In 2023, your coordinate system origin changed based on your alliance color.

For 2024 and beyond, the origin of your coordinate system should always be the "blue" origin. FRC teams should always use botpose_wpiblue for pose-related functionality

 LimelightHelpers.PoseEstimate mt1 = LimelightHelpers.getBotPoseEstimate_wpiBlue("limelight");
      
      if(mt1.tagCount == 1 && mt1.rawFiducials.length == 1)
      {
        if(mt1.rawFiducials[0].ambiguity > .7)
        {
          doRejectUpdate = true;
        }
        if(mt1.rawFiducials[0].distToCamera > 3)
        {
          doRejectUpdate = true;
        }
      }
      if(mt1.tagCount == 0)
      {
        doRejectUpdate = true;
      }

      if(!doRejectUpdate)
      {
        m_poseEstimator.setVisionMeasurementStdDevs(VecBuilder.fill(.5,.5,9999999));
        m_poseEstimator.addVisionMeasurement(
            mt1.pose,
            mt1.timestampSeconds);
      }

Configuring your Limelight's Robot-Space Pose​

LL Forward, LL Right, and LL Up represent distances along the Robot's forward, right, and up vectors if you were to embody the robot. (in meters). LL Roll, Pitch, and Yaw represent the rotation of your Limelight in degrees. You can modify these values and watch the 3D model of the Limelight change in the 3D viewer. Limelight uses this configuration internally to go from the target pose in camera space -> robot pose in field space.