Re: Building a field
 

If anyone is building a field i am sure that i could possibly get a place to host the event and if so that would be in Chaska, MN and we could also bring some of our tech equipment.

Re: Building a field
 

Thanks for the advice. Each year a group of local teams build the field for pre and post season scrimmages. This year the volunteers built the field so it would be easier to modify for future games. Since this one probably won't be torn apart completely, I am trying to put a simple field control in a box to travel with the field. That includes a dedicated laptop as you suggest. So, hopefully with all of your help I can get something put together in the next couple weeks.

I captured the diagram of the FMS layout off of the forum and it seems to be reasonable quality when I printed it(not too fuzzy). If you do have that diagram, I would like to be able to include it in the final instruction manual for setting up the FMS. I don't have photoshop, but can probably find a way to deal with any graphic format.

Thanks again.

Re: Building a field
 

The full-screen image (with the DLink featured) is here:
http://www.team358.org/files/program...te%20setup.jpg

You can also right-click on the picture in my post and save that way too.

I found the original too. Turns out I did it in Powerpoint then took a screen shot of it myself. I put the original here:
http://www.team358.org/files/program...te%20setup.ppt

Just like you're doing, I've got this equipment all packed into a computer paper box with some spares to go to events and demos. Everything is labeled so it doesn't get separated during event teardown. The inventory makes sure it doesn't show up missing one of the 50' cables or something.

Re: Building a field
 

Finally had a chance to test the setup for FMS Light with one robot. It seemed to work like a charm with only two issues. First, I was losing connection to the driver station periodically. These were very short periods of times (maybe a second or two). I didn't have a chance to really trace down the problem. The second issue is that it seems like the DS can be set to 10.xx.yy.5 instead of the 10.xx.yy.4 in the diagram. That makes it easier since that is the setting used normally so it is one less thing to change.

I will have an opportunity to test with 5 robots on Saturday with brand new cables so we will see if the dropped connection is common across all robots and driver stations or just something unique to ours. I am feeling good about the progress so far. Thanks again for the help.

Re: Building a field
 

The DS at .4 is a typo and was intended to be .5 anyway.
Thanks for catching that. Sorry for the mistake though.
I meant to stick to the FRC convention even though other numbers will work as long as you're consistent.
If you decide to go wireless from the Driver Stations, then .9 is appropriate.

FIRST actually has the full FMS IP at .100 if you want to complete the convention.

I don't think this is your problem, but there is a bug in FMS Lite (unless a later version fixed it) that causes a brief field-wide communication shutdown after it's been running about 4.5 hours if I remember right. It appears to be an integer overflow and wrap-around in a variable keeping track of the number of seconds since FMS Lite was started. If it happens during a match all the robots will stop for a second, then start up again.

Re: Building a field
 

Good to know about the glitch in the software. I think I have the latest version of FMS Light which is still a 2010 version. I suspect the problem with periodic communication problems might have been a low battery. I always think of these things when I am at home rather than working on the system. I'll be able to check that on Saturday. After that we will box everything up for an FMS in a box. The problem is that the two students that helped me test everthing won't fit in the box:)

Re: Building a field
 

Have you tried a combination of contortionism and origami?

Re: Building a field
 

Which naturally leads to the question: Do you fix it (assuming you are able) or do you rise a little higher and be like Alan Turing:

http://en.wikipedia.org/wiki/Alan_Turing :)

I have been following this thread with idle interest, not that I have the space or money to build a field of my own. I do have a question, which is more regarding the electronics of a full-staged field than perhaps the "lite" version.

Is it possible to make a "super-network" connecting all the robots, driver stations, and (presumably) programming computers in a competition, then connect this super-network to the field network? All teams have their separate IP number, and I suppose could be separately passworded. When a robot/driver station is to go to the field, the connections are already made and just need the field network to accept the IP, with no rebooting, and disconnect the IP afterwards.

This is no different than an office with separate permissions, but perhaps I'm missing something, not being an IT person.

Re: Building a field
 

I think most of us are like Turing, I just restart FMS Light during the lunch break.
It'd be nicer to fix it, so new users don't have to learn the hard way.

Your off-topic question should be posed in a new thread to avoid contaminating this one. It's not far off-topic, but it does lead into speculation which can run rampant with points and counterpoints.
Off-topic conversations make it difficult to follow the community information provided by a good solid-topic thread. They've become such a pleasure in my old age... :)
I suppose off-season on CD is unlikely to rapidly turn a thread into mush though, and I did my own post wandering.

It's certainly possible to just run everyone in a super-network since that's becoming more common out on the street every day, but probably the increased complexity isn't practical for our intentions.
It would bypass some of the safety features FMS implements and raise new problems, like which laptop is the real driver station when IP addresses accidentally conflict or you want to swap out a misbehaving DS for the next match, or what to do about the increase in collisions/network latency. The networks would have to handle 350+ teams at Championship, typically with the DS, a programming laptop, and scouting computers, mobile roaming devices all active and clammering for attention. Laptops and robots would constantly be adding in, dropping out of the network as they rebooted, went to sleep, were turned off/on.
Nothing more expensive/capable network equipment couldn't handle, but then too, people have emotionally adapted to latency in their personal devices.

None of the masses of laptops or mobile devices would be controlled to avoid network contamination, like how a large IT department would typically protect a business from viral attacks. Network problems aren't even usually malicious, but a result of human error or nature, and in a teaching environment mistakes are common as people are learning.
Some laptops will be connected to the broader Internet, so LulzSec will be driving our robots at times just for the fun of it.

The FMS Light network actually is already partially what you're looking for.
In this network robots are always connected to our field router whether in the pits or on the field, as long as they are powered-up. That doesn't bother us because they aren't saying anything. In the pits the DS wired traffic with the robot goes through the DLink, but isn't broadcast to the field router.
If we extend our field network to include the DS's connecting wirelessly instead of via Ethernet, then we actually will have everyone connected and jabbering through the field router all the time. No priority is given to the field robots though through the noisy traffic.
Depending on the size of the off-season event we may never see a problem with network traffic clogging and slowing the robot response, but we need an IT network person actively monitoring the traffic to be sure. (There, that tied it back into the thread topic.)

Re: Building a field
 

The final update -- Multiple robots connected successfully to FMS Light and the connections appeared to be stable. On to other projects.

Re: Building a field
 

Thanks for the update. :)