[Levansic]

We are having a similar problem, at least in that we step through our sequence, but our drive does not turn off when we give it a zero value at the end. This is a huge problem.

One thing that is different with our code, is that we use periodic tasks to drive, so we only set global variables in our sequence. We are planning to just disable our drive motors at the end of the sequence, and re-enable them at the start of teleop. We may have a bug somewhere in our periodic task code, so that is going to get a lot of looks.

[Update] - found our problem in our periodic tasks. Had an errant write (overwrite) of the global that we set in autonomous.

Of course there are other caveats. We use CAN and get some odd errors from the WPI library for CAN when we read the state of limit switches. We were dropping some packets and maxing out the CPU for no apparent reason in autonomous, which may have been contributing. We addressed the CPU with optimizing some code that is called often. We will be eating the errors to address some latency issues, as the WPI code gives the correct output in spite of the errors. If none of this fixes the problem, who knows.