Wrong Field Measurements and Bad FMS

[sarahg2543]

Hello,
My team is competing at the Colorado Regional this year and they told me that the field measurements were wrong by an inch for the colored zones, which really affected our autonomous. We weren't the only ones who complained. Not to mention, the tape was diagonal, not straight. The field crew told us that it was the robot who needed to adjust their autonomous, but we just eye-balled positioning our robot by an inch. Did anybody else hear about this? I think it's pretty serious when it is affecting the autonomous of teams, especially since it can really make a difference in points. I also know that the practice field measurements were wrong.

Another issue that we encountered was that the FMS tended to malfunction, especially in the red alliance driver station. I saw a lot of robots lose their communication. The field crew said that it was due to the robot most of the time instead of replaying the match.

This is jut to point out that there are still some things that need to be fixed in fields for the next week of competition.

[Alex Cormier]

There's always a field tolerance. It's in the manual each year.

[EricH]

Originally Posted by Alex Cormier

There's always a field tolerance. It's in the manual each year.

Also, you're allowed onto the field to measure it starting an hour before practice matches per the Tournament rules.

[alectronic]

Originally Posted by sarahg2543

Another issue that we encountered was that the FMS tended to malfunction, especially in the red alliance driver station. I saw a lot of robots lose their communication. The field crew said that it was due to the robot most of the time instead of replaying the match.

Do you have any evidence that the "FMS Malfunctioned"? The field crew can see the log information (or most of it) from your robot, which helps them determine if there was indeed a fault or if everything looked good. The field crew is out there to make sure things are working correctly, presumably they would have investigated and determined it wasn't FMS before they made any decisions.

[Alan Anderson]

Originally Posted by sarahg2543

Another issue that we encountered was that the FMS tended to malfunction, especially in the red alliance driver station. I saw a lot of robots lose their communication. The field crew said that it was due to the robot most of the time instead of replaying the match.

Loss of communication is due to the robot most of the time. If it isn't the robot, then it's almost always the Operator Console. If it looks like one particular alliance station spot has problems for several matches in a row, it's most likely just a coincidence.

I've been a CSA at many events over the past few years, and I'm pretty confident about this. One team's mentor came up to me at the Boilermaker regional and told me their scouting group had identified the Blue 3 position as having a problem, with three robots not working there before their own robot had control issues in their most recent match. I had just come from their pit where I showed the team the loose battery connection on their robot, and I informed him that the three previous robots had suffered 1) an unplugged battery, 2) a shorted power wire on the cRIO, and 3) a disconnected joystick.

[mrnoble]

The practice field was donated by one of the teams (2996) and was close to spec considering it was put together by a bunch of students. We measured the field and tried to adjust our auto to compensate.

[MrRoboSteve]

I was CSA at three events this year. Every communication issue I dealt with was traced to an issue on the robot or with the team's driver station.

The closest we had to a "field" connectivity issue was needing to swap out a DLink.

[Jon Stratis]

I can tell you from experience... these fields stretch. At Lake Superior, the carpet got so bunched up behind the drivers station it was almost a hazard. That comes from the carpet being taped down on the edges, and the field itself expanding as robots bash into it.

A good rule of thumb for any autonomous mode is to make it capable of 6 inches of play or more in any dimension. That'll take care of field dimensions being off a little AND robot position being off a little.

[JosephC]

Do you happen to have any pictures of the diagonal tape lines, that'd be the only field issue I see in your post, and can't really address is without a picture.

[Michael Hill]

To be perfectly honest, 1 inch sounds pretty reasonable. We have real design constraints and tolerances that you have to design your robot around. This isn't the "Perfect Physics world in a vacuum" where all measurements are spot on. They're close, but not perfect. The field is [mostly] held together with velcro, pins and zip ties. What do you expect?

[MarcD79]

As Field Supervisor at the Groton & Southington District plus Field reset at the Hartford District in Connecticut, I can tell you that the field definitely grew an average of 2-3 inches resulting in the classic folds on each end. In Groton the carpet was taped down on a brown craft paper. In Southington & Hartford the carpet was taped down on a vinyl rolled-out layer, which did stretch. As for being off an inch in the colored zones, I saw many robots shooting on the fly, so that "1 inch" was a moot point. If you are relying on being within 1 inch so you can shoot your ball, that takes too much time to line up & results in longer cycle times than someone who can just shoot as they fly down the field.

[Jay O'Donnell]

Originally Posted by Michael Hill

To be perfectly honest, 1 inch sounds pretty reasonable. We have real design constraints and tolerances that you have to design your robot around. This isn't the "Perfect Physics world in a vacuum" where all measurements are spot on. They're close, but not perfect. The field is [mostly] held together with velcro, pins and zip ties. What do you expect?

I expect people to put tape in the right place. Seriously an inch is honestly a pretty big deal for some teams. Why can't we put tape in the right spot? It's not that hard.

[RufflesRidge]

Originally Posted by Jay O'Donnell

I expect people to put tape in the right place. Seriously an inch is honestly a pretty big deal for some teams. Why can't we put tape in the right spot? It's not that hard.

The overall dimension of the field is specced as +4.0/-0in. and the tape is specced to trisect this area. This means that +1.33in. should be the tolerance for the size of each zone. Additionally the field and carpet stretching as mentioned by other posters above may cause additional deviation from this measurement that is very difficult for FIRST/event staff to do anything about.

If 1 inch honestly effects your autonomous:
1. You have a pretty solid PID running, it should be trivial for your team to tweak the value to fix your auto
2. It is unlikely your robot will score many goals in teleop this season unless you are using automatic positioning. A driver cannot achieve inch accurate positioning from 30+ feet away. Even with auto positioning any defense at all will cause a deviation of substantially more than an inch.

[Bryan Herbst]

Originally Posted by Jay O'Donnell

I expect people to put tape in the right place. Seriously an inch is honestly a pretty big deal for some teams. Why can't we put tape in the right spot? It's not that hard.

We can and should expect tape to be placed in the right spot. Unfortunately I think that part of this thread got lost amongst the unfounded FMS complaints.

It also wasn't clear whether the tape was incorrectly placed from the start, or whether the tape was placed correctly then appeared to be incorrect after a weekend full of matches.

This is an easy problem to deal with appropriately. As per section 5.5.7 of the manual, the field will be open for measurements for one hour at some point prior to practice matches starting. Go out on the field, take a look at the small details that you know will affect your robot, and point out any errors that are outside the tolerances to the Field Supervisor and/or FTA.

[Doug Frisk]

Originally Posted by sarahg2543

Another issue that we encountered was that the FMS tended to malfunction, especially in the red alliance driver station. I saw a lot of robots lose their communication. The field crew said that it was due to the robot most of the time instead of replaying the match.

Not that the FMS can't be temperamental, but as has been said, it's almost always the robot. I spent 3 weekends staring at the FMS console this year and in that time:

I saw well over 100 cRIO reboots.
I saw at least 3 dozen main breakers blow.
I saw a couple dozen situations where the radio or cRIO came unplugged during a match.
I saw plenty of drivers stations freeze.
I saw drivers unplug their joysticks or the drivers station Ethernet connection mid-match.
I saw a couple of robots running dual 1080p cameras kicking up around 15 megabits of bandwidth.

The closest I saw to an actual FMS (communications) issue was that the trip time from driver's station to robot spiked across all 6 stations to 120 milliseconds for a second or two then dropped back under 5 milliseconds. (despite my own concern, no robots dropped and the teams didn't notice anything during the match) Not to say that the FMS can't be an issue, but there are a lot of other places to look first.