Driver station disconnect issue?

[joeltruher]

Hi,

I'm a mentor for FRC Team 100, and I'm looking for someone at NI responsible for the driver station code, specifically the part that governs the safety-disable feature involving a heartbeat from the driver station to the RoboRIO.  In an effort to improve "liveness" of the disable feature, the level of "false positive" disconnects went up in the 2024 release.

There's been some discussion of this issue over the past year or so here:

https://www.chiefdelphi.com/t/internal-issue-with-print-and-error-tags-driverstation-randomly-disabl...

I had assumed that someone at NI was working on refining this behavior to reduce the level of false positive disconnects, but now I'm wondering if that's actually true.

Can anyone help?

Thanks!

[Greg_McKaskle2]

That is a true assumption. The beta program has just started and we intend to tweak the code and values during the beta. I haven't visited the forums yet due to work deadlines. 

[joeltruher]

Oh, that's terrific, thanks for getting back to me!  Is there any way I can help?  Or any way to measure more precisely what's happening?

[joeltruher]

here's another observation

https://photos.app.goo.gl/cswdjNeFt8mzShLg7

flipping between "teleop enabled" and "no robot communication" with a period of about 0.5s.

this was using an FMS.

has anyone seen this before?

at the same event, we also had long disconnections, though neither the laptop cpu nor the roborio cpu were more than 10 percent busy.