Detecting Path Completion

This page serves as a guide for determining when the robot has finished following a Path or PathChain object and is ready to follow the next one.

In general, the ideal way to determine this is through checking !follower.isBusy() which would return true when the path is complete and false otherwise. The method follower.isBusy() becomes false after the robot is done correcting at the end of the path. If you don't want to wait until the follower is done correcting, you can instead use follower.atParametricEnd() along with a heading check (i.e., follower.getHeadingError() < follower.getCurrentPath().getHeadingConstraint()) to detect that the path is complete.

Sometimes, calls to !follower.isBusy() may stay false even when the path appears to be complete. This generally occurs when either the heading or translational end constraints aren't met, preventing the Follower instance from thinking the path is complete. This issue typically happens almost every time the autonomous is run, so it becomes easily apparent when this will be a problem. The problem occurs when the secondaryTranslationalPIDF or secondaryHeadingPIDF aren't powerful enough to correct the robot to the endpoint exactly, but they can't be made stronger without impacting path-following performance. In such cases, there are two solutions: you can either lower the bar for the PathConstraints, which can be done easily in the PathBuilder, or use follower.getVelocity.getMagnitude() < follower.getCurrentPath().getVelocityConstraint() along with a check to the distance of the robot's current pose to the last point in the path. A similar constraint can be put on heading when necessary, for example, follower.getAngularVelocity() < 0.055 with a more lenient check on the heading error.

A good autonomous can sometimes even have failsafes on whether the follower thinks the path is completed, for example, tracking the amount of time the robot moves for a set of preloads and having the follower automatically assume the path is completed a given amount of time after the robot begins the path. This can be done with either a basic timer or a TemporalCallback.