My team, as well as others have run into this problem. Something on your robot loses power, and you lose communication. No one can really tell you what happened, because they don’t know.
This is especially frustrating. I have done testing at our shop by unplugging the radio and plugging it back in, unplugging the drivers station and plugging it back in, even removing the power leads from the CRIO for a brief moment, then plugging it back in. In all situations the robot regained communication with the KOP router within 30-40 seconds.
How can that be? As anyone who has had this happen to them can tell you, the robot doesn’t regain communication…
How can a team debug this issue, and expect to actually find the problem when it is impossible to recreate the problem without an actual field. There must be something that the FRC field is doing that our KOP routers are not, that disallows them to reconnect to the network. I want to know what that is, so that we can either recreate it, or possibly talk someone into disabling it.
Just curious if anyone knows the answer/wants to chime in.
I assisted a team at the DC Regional whose robot had consistent communications dropouts on the field, but which had never had problems at home. Moving the WGA out from underneath a Jaguar made everything work properly.
If your robot’s radio is in a bad location, but your testing environment is giving a strong enough signal between the robot and the router, you might never see anything break.
At the Michigan State Championships we were seeing our robot drop out light a lightbulb from just driving around. It was caused by an 1/8th inch aluminum L bracket holding the radio. When we drove around apparently it caused enough vibration to disrupt the signal.
As soon as this information was found we immediately stopped using the bracket, and all seemed to be fixed.
With our new robot at MARC we died twice, both times after bumping into a wall, both times were very, very minor crashes. 1/10th of that seen through the rest of the competition with collisions of robots.
I don’t think this is a configuration of the components. I am sure we have a connection problem somewhere, we have already found two. I am still curious why the FRC field software will not let your robot reconnect to the network after losing comms.
I did see a couple robots this year that had the radio mounted in such a way that the reset button would be pushed if the robot had collisions in certain conditions.
I just had a thought: maybe the robot goes into watchdog mode when it loses communication while under fms’s control and cannot get out of it but can get back out of watchdog while its on an open wireless connection.
or maybe when it loses communication the fms program reacts by shutting the robot comm down and putting it in watchdog as a ‘safety precaution.’
both are possible i guess.
hmmm… could be a special form of hey-i’m-gonna-really-screw-you-guys-up watchdog that won’t let it communicate. We’ve never had this problem (thankfully) so i’m not much help i guess
i most definitely agree, however it’d probably take a LOT of time to go through all the possibilities. then again maybe they already fixed it for 2010.
it may be in the way the bridges are set up in the competition
the bridge on the robot is a router, and the fms connects to it, (thats why there are 6 antennas on the wifi box on the field)
but in the shop the bridge is solely a wireless to ethernet bridge and you have a router
the fms bridges may only try to connect once, so if you lose communications it won’t try to reconnect (fms)
I believe you have to reset your Driver Station to do this. Takes around 20 - 30 seconds. Obviously a big deal in an official match but it can be done.
I have never directly had this issue but it seems certain teams had the issue repeatedly at regionals. I certainly hope this doesn’t plague 47 at IRI D:
It’s very confusing to other people on the message boards when you make things up that are incorrect and assert them as fact. Please don’t do this.
The wireless bridge on the robot is a bridge regardless of if you’re at home or at a competition. The device at the side of the field at competition is indeed an access point that the bridges on the robot connect to. The reason there are 6 antennas is because that access point has both 2.4GHz and 5GHz. Each band uses 3 antennas because it employs MIMO technology.
…and it won’t. The only time you will see watchdog written to the DS screen is when the user watchdog is killed, and if you are seeing this you are connected to the robot.
Watchdog doesnt stop your robot from talking to the controller, and if that was the case, than I would assume that you would lose connection, watchdog initiated, reconnect, display watchdog.
