PDA

View Full Version : Mech Warfare 2010 Draft Rules



lnxfergy
07-29-2009, 12:33 PM
>> Original Post

Ladies and Gentlemen,

After careful deliberation, and multiple edits, I am happy to announce the official draft of the 2010 rules. Multiple clarifications and updates, related to experience from our trial year, have been incorporated into the rules.

This thread is to serve as a public comment space. The official comment period is from now until August 31st, 2009. Andrew and I will take all suggestions into account, however, we cannot guarantee that they will be incorporated into the rule set. A final rule set should be posted sometime in September, along with finalized details on the scoring system, including purchasing information.

-Fergs

>> April 6, 2010 Update

The final rules are finally posted here:


Mech Warfare Rules
2010 Edition

JULY 2009 DRAFT, CHANGES AND CLARIFICATIONS FROM THE 2009 RULES ARE IN RED.
FINAL CHANGES FROM JULY 2009 ARE IN BLUE

Mech Warfare is a robotics competition. Our goal is to create a real-life robotic combat competition that mirror the scenarios found in sci-fi universes such as Battletech, Warhammer 40k and Armored Core. Competitors will build 1/24 scale armed robots which they will pilot through a wireless first-person POV system.

Section I - Leagues of Play

There are three leagues of competition:


The Airsoft Classic league is open to all walking robots with 4 or less legs.

The Airsoft Hexapod league is open to all 6-legged robots. It is intended as an easier entry point into Mech Warfare.

The Hardcore league is open to all walking robots with 6 or less legs.

Section II – Mech Construction
All competitors are subject to final approval by Mech Warfare officials. Each competitor must be presented to an official for a Safety and Technical Inspection before competing, and must be re-inspected after any major alterations. A mech may be rejected from competing if it is deemed to be unsafe for human bystanders or not in the spirit of the event. All mechs should be designed within the spirit of the game. Do not try to 'gimmick' yourself to a win. If you have a question about a particular part of your robot, ask the event organizers beforehand.


Mechs are to be true walking robots. Legs must be servo/actuator driven. No cam-driven, wheeled, or treaded configurations.
Mechs may have up to 4 legs. (6 in the hexapod league).

Average mech size is expected to be between 8” and 18” tall. No robot may be taller than 36”.

Mechs may be remotely-operated or autonomous.
Pilots may only view the match through their first person POV camera mounted on their bot, further:

Pilots are not allowed to view the arena or match directly.
Cameras should be mounted roughly in the center of the mass of the robot, where the 'cockpit' would be. The intention is to simulate piloting the Mech, not having cameras on your guns mounted to extensions so you can fire around corners without fear of being shot. This would fall under the 'Gimmick Clause'.
Wireless IP Cameras are recommended. Non-wifi cameras will not work in the high RF interference arena.
Spectators will be able to view the full arena and match,however they should not provide hints, tips or assistance to pilots.


Mechs should not separate, or leave pieces of themselves behind, especially beacons or debris that would inhibit other competitors.
Each pilot should send the following information to the event organizers at least 4 weeks before the event so that competitor information can be added to the website, and on-site promotional information:

Mech Name, Pilot Name
Basic outline of how the mech is constructed, what parts it uses.
A picture of the mech, on a neutral background, at least 800x600 pixels



Note that weight restrictions have been removed. Once participant density allows, separate lightweight and heavyweight classes will be formed.

Section III – Weapons Systems
There are two separate sets of weapons rules, one for each of the Airsoft or Hardcore classes. Mechs may be outfitted to run either class, however, only guns legal for the current match may be loaded and active.


The Airsoft class allows the use of Nerf weapons and electric Airsoft guns. Guns are to use standard Airsoft 6mm plastic BB ammo. While there is no hard limit on your gun's muzzle velocity, it must be low enough so as not to break the mesh walls of the arena. When outside the competition arena, all guns must have a physical barrel lock in place which prevents BBs from being fired.
The Hardcore class will be offered based on availability of a fully enclosed combat cage. The Hardcore class allows use of much more powerful weapons such as CO2 powered rifles, micro class rockets, and flamethrowers.
In all weapons classes, any mech that is capable of shooting without explicit human input must have a verified remote kill switch and a visual indicator that they are "armed". We will not have Skynet go live on our watch.
In all weapons classes, weapons designed to interfere in any way with an opponent's camera or wireless control are strictly forbidden.

Section IV – Arena


The airsoft arena is approximately 15'x15'. The hardcore arena is 8'x8'.
The walls of the arena will be non-transparent to at least a height of 12”.
Buildings, averaging two feet tall, will be provided for cover. Building layout will be consistent throughout the event for ranked matches, however layout may not be finalized until the event. Exhibition matches may have varied building layouts at coordinators discretion.

No 'street' will be less than 36” wide.
The corners will be masked off with 3'x3' squares taped on the floor denoting the starting positions for mechs.

All sizes are approximate. Your mech must be able to deal with any small changes in these dimensions.

Section V – Match Rules
A match consists of two mechs facing off in an arena, trying to decimate one another's Hit Points (HP). The winner is the mech with the most HP left at the end of a match. The Scoring system consists of target plates and a transponder unit which wirelessly relays information back to a base station.


There will be a bulletin board denoting when matches are to be held. The event organizers will give 5 minute and 1 minute warnings. If you mech is not ready for a match, it will forfeit that match. Builders will be give one 2-minute extension to use throughout the weekend.
Each Mech will start a match with 15 hit points (HP). Remaining HP will be reduced as the transponder unit registers hits, or when penalty hits are assessed. Starting HP may be raised, if at the end of any day the average match time, not including forfeited matches, is less than 8 minutes.
Mechs will start in opposite corners of the arena. All of a mech's feet must fully cross the corner tape before it can score a hit on it's opponent.
Competitors reduce an opponent's HP by scoring hits on an opponent's target plates
The scoring system will not score more than 1 hit per second, regardless of how often it is hit.
A collision or knockover will score as a hit if the scoring transponder detects it. If your mech is unable to right itself, the match will be paused while the mech is assisted and penalized an additional hit point of damage. The clock will not stop during any assistance.
If a mech does not move for 20 seconds, it will be assessed 1 penalty hit point. If a mech does not move for 60 seconds, it will be determined 'dead' and forfeit the match. Panning/tilting of turrets does not count as movement. This rule is intended to encourage mobility and prevent American Civil War Era fighting strategies.
A match ends when either one of the mech's has it's HP reduced to 0, or when the match clock runs past the 12 minute limit. The mech with the higher HP at the match's end wins.

Section VI – Scoring Transponders

The scoring transponders and target plates are distributed by the Mech Warfare organizers, overall cost to participants should be less than $100 per entry.


Target plate specifications:

Target plates are 3”x3”, weighing approximately 20g, and must be purchased from the event organizers. The plates should not be altered, other than to apply velcro or similar to the back for attachment to the mech. Due to technological constraints, plate design may change from time to time, however event organizers will strive to be consistent from year to year with the plate design.
A Mech must carry as many target plates as it has legs. Bipeds require one front and one rear target plate. Tripods and Quadrapods must have at least one scoring plate visible from any horizontal angle.
Target plates must be reasonably mounted on a mech,with their entire face located between 2” and 22” off the ground, and not obscured by any limbs. Use common sense when choosing a mounting location, and keep in mind the spirit of the game. If you are firing on an opponent, they should be capable of hitting your scoring plates.For instance, if a biped has it's camera and guns on a turret, the target plates must also be on the turret.

In order to allow autonomous bots, and those using visual tracking, competitors may bring a visual fiducial of any color which may be applied to an opponent's target plates using tape of any color. Fiducials should be no bigger than 3"x3". As these fidicuals may become damaged during use, it is suggested to bring a decent quantity of them with you.


Scoring transponder unit specifications:

Transponders are approximately 2.4”x2.4” weighing approximately 25g.
The unit requires a power connection capable of providing 7-16VDC at up to 200mA.
The unit will send out a 200ms high pulse each time it reports a hit. This can be tied into your robot to allow your control solution to register hits.
The unit will be programmed with the appropriate firmware and ID information when distributed. The event organizers reserve the right to require firmware upgrades at the competition, so be sure that the In-System Programing header on the transponder is easily accessible.


Scoring displays will be set up in locations visible to both competitors and spectators.


Section VII – Record of Changes


June 22, 2009 - Document created from 2009 rule set

June 23, 2009 - Slight changes, notes added. Removed weight limit (we will have weight classes some day).
July 29, 2009 - Final highlighting changes before release to general public.

December 7, 2009 - Revision for final release.

April 6, 2010 - Actually posted (yay!)

Adrenalynn
07-29-2009, 12:41 PM
I haven't read it, but I'm just posting to note: Stickied!

CogswellCogs
07-29-2009, 01:15 PM
These look really good. Thank you for putting them together.

I noticed that the meelee rules are now gone. Will meelee and kung-fu attacks be allowed ?

mannyr7
07-29-2009, 04:14 PM
Section II - 3 needs clarification:


Auto-aiming and lock-on systems are allowed. Automatic firing is not allowed, unless the mech is completely autonomous. Autonomous mechs must show a verified remote shut-off, we will not have Skynet go live on our watch. All our weapons are automatic, in the sense that one trigger pull results in a continuous stream of rounds being fired. I take it, by automatic, you mean the autonomous robot chooses when to fire.

Adrenalynn
07-29-2009, 04:42 PM
The weapons are not "automatic firing" - that's a semantics game. They're automatic cycling, not automatic firing if we really must play that game...

mannyr7
07-29-2009, 04:50 PM
Sorry, seemed ambiguous to me. 8 years in the Marine Corps, "automatic fire" means something entirely different to me.

DresnerRobotics
07-29-2009, 04:58 PM
It's a good point, we'll clarify.

Adrenalynn
07-29-2009, 05:02 PM
We need a good smiley for "biting tongue". ;)

mannyr7
07-29-2009, 06:15 PM
http://forums.trossenrobotics.com/gallery/files/2/8/4/0/16x16_robot-bites-tongue.gif No worries! ;)

Adrenalynn
07-29-2009, 06:19 PM
Perfect.

And thank you for your service, BTW. Semper Fi! [From a 34 year USMC-Brat]

gdubb2
07-29-2009, 09:36 PM
Section IV: #3

[quote]Mechs will start in opposite corners of the arena. Both of a mech's feet must fully cross the corner tape before it can score a hit on it's opponent. [QUOTE]

Perhaps instead of "Both of a mech's..." it should be "All of a mech's..." to account for more mech's with than 2 feet. More semantics..

Gary

robologist
07-30-2009, 01:12 AM
Regarding weapons systems, how would a competitor know if their hopped up Airsoft gun will pierce the mesh while they are developing their bot at home?

Of course, I also don't know how muzzle velocity or kinetic energy of a round might be measured for alternate rule limitations.

Was there anyone trying to improve their Airsoft performance? Or homebrew something?

Adrenalynn
07-30-2009, 01:34 AM
Sienna originally suggested a joules measurement, and I think we might want to discuss getting that back in there.

elios
07-30-2009, 01:52 AM
maybe death match....? like four mechs in the arena at any one time, for a certain period of time. 3 minutes? to score as many hitpoints as possible on other mechs? once the time is up, they power down and ranked 1st, to last based on how many hits they had?

just throwing it out there... might not be possible with conflicting wifi etc(?) with the webcams, if many people have the same type. or is there like a "crystal" which can be changed?

cheers,

DresnerRobotics
07-30-2009, 02:23 AM
Regarding weapons systems, how would a competitor know if their hopped up Airsoft gun will pierce the mesh while they are developing their bot at home?

Of course, I also don't know how muzzle velocity or kinetic energy of a round might be measured for alternate rule limitations.

Was there anyone trying to improve their Airsoft performance? Or homebrew something?


Well, we have known goods. Pretty much any electric toy gun from Walmart is going to be fine. The tank guns are also fine. Anything more powerful than that, you're going to need to test beforehand or risk it and test at the competition before competing. Once I finalize the revised arena design, I'll publish where I'm getting the materials (its cheap) so people can order it if they like and do their own testing. I dont think it needs to get too complex in terms of measuring velocity, as long as it doesn't penetrate the mesh it will be fine.

TrevorBrummel
07-30-2009, 09:03 AM
Here's a question: Are directed energy weapons allowed? I.e. could I mount either a full-fledged, plastic-burning green laser for hardcore matches or something like a red one to blind cameras with in the airsoft matches?

jes1510
07-30-2009, 09:47 AM
Nothing is allowed to interfere with the cameras or directly damage the mech. The idea in hardcore is not to damage the mech. The idea is to allow a bit broader spectrum of weapons. I would be really unhappy if I spent all the time building a mech and you destroy it with a saw.

On a related note, do you really want to fire a high power green laser inside a clear lexan arena with children looking through the walls?

CogswellCogs
07-30-2009, 10:14 AM
Nothing is allowed to interfere with the cameras or directly damage the mech.

I wondered about trying to blind the opponent's camera with bright LEDs. Not enough for permanent damage, just to make it ineffective. I've also considered that it may violate the 'gimmick' or 'spirit of the mech' ideals. What do you think ?

lnxfergy
07-30-2009, 10:37 AM
Section II - 3 needs clarification:

All our weapons are automatic, in the sense that one trigger pull results in a continuous stream of rounds being fired. I take it, by automatic, you mean the autonomous robot chooses when to fire.

We do mean "fire without pulling the trigger, ever." Will be clarified.


Perhaps instead of "Both of a mech's..." it should be "All of a mech's..." to account for more mech's with than 2 feet. More semantics..

GaryYep, obviously I was thinking mainly about bipeds. This is our so-called "anti-turret" or "no german-pillbox" rule. The wording will be changed.


Regarding weapons systems, how would a competitor know if their hopped up Airsoft gun will pierce the mesh while they are developing their bot at home?

Of course, I also don't know how muzzle velocity or kinetic energy of a round might be measured for alternate rule limitations.

Was there anyone trying to improve their Airsoft performance? Or homebrew something?

Frankly, I'm not sure there is much to be gained by hopping up the velocity of the guns. Maybe the BB will fly a little straighter, but let's face it, at most you are shooting 12 or so feet, and can you even see the target at 12 feet when using a 320x240 camera? Higher rate of fire would be more useful. We're mainly concerned about safety with this. I would reckon that all of the possible guns that a mech can actually carry, if left stock, will be way below any threshold for safety.


I wondered about trying to blind the opponent's camera with bright LEDs. Not enough for permanent damage, just to make it ineffective. I've also considered that it may violate the 'gimmick' or 'spirit of the mech' ideals. What do you think ?

During the discussion of the rules last year, it was noted that toying around with the cameras was forbidden. It's definitely gimmick material. Somehow, it didn't make it into the draft rules... you can be certain that it will be in the final rules.

-Fergs

mannyr7
07-30-2009, 03:47 PM
During the discussion of the rules last year, it was noted that toying around with the cameras was forbidden. It's definitely gimmick material. Somehow, it didn't make it into the draft rules... you can be certain that it will be in the final rules.

-Fergs

Better include radio jamming, EMPs, water/acid squirt guns, deployable drones, black hole generators, and all the other nonsensical ideas that came out last time. ;)

ScuD
07-30-2009, 04:15 PM
Darn.. and I just finished finetuning the anti-matter cannon.. :(

mannyr7
07-30-2009, 05:06 PM
If you got a functioning unit and a mech capable of hauling it around, I say enter it in the hardcore class! My post was directed more toward the gimmicky loophole what-ifs. :)

TrevorBrummel
07-31-2009, 06:11 PM
Here's another question: will robots be allowed to climb or enter the structures/buildings in the arena?

mannyr7
07-31-2009, 08:11 PM
Last discussion about building design was cardboard boxes with photocopied prints of real buildings pasted/velcroed on, so no. No entering buildings. Some discussion has been thrown around about destructible buildings, but still too early to tell if this would be feasible for transportation, construction costs, cleanup, and safety/reliability. I know I shouldn't even go here :D, buuutttt... i suppose if you had jumpjets you could attempt to land on buildings. :p

jes1510
07-31-2009, 08:26 PM
Personally I would say that if you can get on top of a building then go for it. If you can do it cheaply then patent it and sell the design for a bazillion dollars

DresnerRobotics
07-31-2009, 08:37 PM
Here's another question: will robots be allowed to climb or enter the structures/buildings in the arena?

A better question is if any robot capable of competing in this competition would even be able to climb a 2 foot structure.

Indulgence in the 'what ifs' is an unfortunate downfall of progression in one's project.

TrevorBrummel
08-01-2009, 09:13 AM
Ok. I just wondered because I saw this:

http://mech-warfare.com/images/arena1.jpg

I thought that there could be the potential for robots getting on top of structures, so I thought I'd ask if it was allowed under the rules.

darkback2
08-01-2009, 09:38 AM
I think early on we had a lot of ideas that reality has dispelled...at current the arena has to be something that we can put together and take apart. It has to be relatively inexpensive, and it has to meet the requirements for mech warfare. The buildings did actually become an issue of sorts. We originally talked about foam buildings that would be destroyed by pellets...cool in theory...but expensive because the buildings would need to be built for every event.

There was talk about stackable buildings, but these would be relatively expensive up front, and would have to be shipped from place to place...Also, all of the buildings would be the same.

For now at least, cardboard boxes are cheap, and available anywhere. I'm going to try to get some store fronts printed up that are big enough to actually cover the sides of the boxes. I took a bunch of pics of store fronts in LA and will take some in New York in the next couple of weeks. 8 boxes was the last count, so 32 pictures at 12 inches wide...good news is if we do it right the pictures could last a really long time.

Now there is nothing to say that we couldn't make some cardboard pill boxes...

Maybe a mod could move this discussion to the arena discussions thread?

DB

Connor
08-01-2009, 01:30 PM
I think early on we had a lot of ideas that reality has dispelled...at current the arena has to be something that we can put together and take apart. It has to be relatively inexpensive, and it has to meet the requirements for mech warfare. The buildings did actually become an issue of sorts. We originally talked about foam buildings that would be destroyed by pellets...cool in theory...but expensive because the buildings would need to be built for every event.

There was talk about stackable buildings, but these would be relatively expensive up front, and would have to be shipped from place to place...Also, all of the buildings would be the same.

For now at least, cardboard boxes are cheap, and available anywhere. I'm going to try to get some store fronts printed up that are big enough to actually cover the sides of the boxes. I took a bunch of pics of store fronts in LA and will take some in New York in the next couple of weeks. 8 boxes was the last count, so 32 pictures at 12 inches wide...good news is if we do it right the pictures could last a really long time.

Now there is nothing to say that we couldn't make some cardboard pill boxes...

Maybe a mod could move this discussion to the arena discussions thread?

DB

Just a note, we didn't do it this past time, but, those boxes could be placed on thier side to be a 2' wide by 1' tall.. That would be cool too.

Thanks, Connor

Adrenalynn
08-01-2009, 02:10 PM
There's no rule against bots climbing onto the structures. Or climbing up the netting. Or dematerializing, traveling through subspace, and reappearing behind the opponent. From my perspective those are all fair game.

I would respectfully submit for consideration that one build a 'bot that can successfully accomplish the basic task of walking around via wireless control with wifi camera and accurately fire a basic airsoft tank weapon, and do so for 15-20mins reliably, prior to considering the above though. I think it has been pretty well established that the latter is more involved than it would initially appear.

MSK Mech Commander
08-04-2009, 03:00 PM
I'm going to try to get some store fronts printed up that are big enough to actually cover the sides of the boxes. I took a bunch of pics of store fronts in LA and will take some in New York in the next couple of weeks.
DB

Maybe as the event gets bigger you could get pictures of famous cities from around the world, like Tokyo, London, or maybe even Venice?;)

robologist
08-05-2009, 03:35 AM
How would the pictures from Google Street View work? Too fuzzy?

mannyr7
08-13-2009, 01:10 AM
Regarding mech construction: My mech has a pan and tilt head with weapons and vision tied together. Is it enough to mount the target plates, unobscured, on the front and back, fixed part of the torso? Or do they have to be mounted to the moving turret, so that if I'm looking and shooting, I can be engaged from the same direction? I'm not trying to skirt the letter or intent of the rules. I am just trying to build something truly mech-like, out of video games and would prefer my head not look like a 3" x 3" cube with guns. Similar to this:
http://forums.trossenrobotics.com/gallery/files/2/8/4/0/madcat1.jpg

lnxfergy
08-13-2009, 09:55 AM
The general idea is: if you are able to shoot at someone, your target plates should be lined up so that they could return fire if they were facing at you. Therefore, that mounting option you show is outside the rules, unless you can demonstrate that there is a physical stop on the turret that allows it to only turn about 15-20 degrees in each direction (which would be enough to make sure the target plate is visible whenever you are gunning down someone). It would be important to make sure that when it turns, the guns also don't obscure the plates.

-Fergs

darkback2
08-13-2009, 01:54 PM
Oddly I think this presents quads with an odd advantage. A quad can technically have its target plates mounted on the body with a separate turret because with 4 target plates a target plate will always be visible. Both Squidword and Charlie had that configuration...I suppose I could change it for next year if need be.

I'm wondering if the rule should be something more like if you can see the opponent then they can shoot you...tying the target plate to the camera and not to the gun.

This seams more logical given that the target plates are supposed to mimic the location of the cockpit.
DB

Advantage is probably the wrong word.

mannyr7
08-13-2009, 04:18 PM
Thanks for the clarification Fergs. This should be added to Section 1: Mech Construction. Oh and that pic was a bad example. My guns will be level with my "cockpit", so the guns would not block my plates while turned. But, I got ya. Back to the drawing board... :cool:


Oddly I think this presents quads with an odd advantage. A quad can technically have its target plates mounted on the body with a separate turret because with 4 target plates a target plate will always be visible. Both Squidword and Charlie had that configuration...I suppose I could change it for next year if need be.

I'm wondering if the rule should be something more like if you can see the opponent then they can shoot you...tying the target plate to the camera and not to the gun.

This seams more logical given that the target plates are supposed to mimic the location of the cockpit.
DB

Advantage is probably the wrong word.

I was thinking this as well. The target plates on a quad could present up to 45 deg deflection to an opponent. I don't know what kind of sensor we'll end up using, but this could result in a non-hit.

lnxfergy
08-13-2009, 04:42 PM
Oddly I think this presents quads with an odd advantage. A quad can technically have its target plates mounted on the body with a separate turret because with 4 target plates a target plate will always be visible. Both Squidword and Charlie had that configuration...I suppose I could change it for next year if need be.

I'm wondering if the rule should be something more like if you can see the opponent then they can shoot you...tying the target plate to the camera and not to the gun.


I don't see how this is an "advantage" other than the fact that you could put less weight on your turret for a quad. We've already accepted that building a biped is harder/more expensive, and I only see this as an extension here.

As for viewing or shooting, the guns should be lined up with the camera anyways, so the two are one in the same in my book. I can't think of any particular way of building a bot where the camera/guns are not in the same direction, and that doesn't get hit with the "gimmick clause".


Thanks for the clarification Fergs. This should be added to Section 1: Mech Construction.

Section V, parts 2 and 3 pretty well cover all of this, I've underlined for emphasis.





A Mech must carry as many target plates as it has legs. Bipeds require one front and one rear target plate. Tripods and Quadrapods must have at least one scoring plate visible from any horizontal angle.
Target plates must be reasonably mounted on a mech,with their entire face located between 2 and 22 off the ground, and not obscured by any limbs. Use common sense when choosing a mounting location, and keep in mind the spirit of the game. If you are firing on an opponent, they should be capable of hitting your scoring plates.



-Fergs

robokoi
08-18-2009, 06:32 PM
I like how the rules are shaping up and look forward to building a bot. Given the addition of a hexapod league, this seems to be missing from the current set (both here and on the website), e.g. sec 1.2.

Couple questions:
- In sec 1, are the Safety and Technical Inspection regs listed anywhere for what we have to follow to be allowed to compete? Not sure if this is meant to be from the safety threads here or something specific to RoboGames.

- On the arena walls, are they really mesh? As in a screen with holes? Is it possible to list the material and/or provide samples to participants (cost or otherwise) for safety testing?

- Sec 4.3, is the 3x3 no-hit area tied into the scoring system somehow? It might be easier to just guarantee that the bots will not have visual contact at the start of the match, either though buildings blocking the view or by having them face away from the center?

- On the target system, is that $100 for a 4-board system w/ transponder? I'll try to dig up the current thread for targets, but it's 3" x 3" x what? Are there any requirements on the placement of the transponder?


ok, more than a couple questions, I guess.

lnxfergy
08-18-2009, 06:44 PM
I like how the rules are shaping up and look forward to building a bot. Given the addition of a hexapod league, this seems to be missing from the current set (both here and on the website), e.g. sec 1.2.
Hexapod league was announced after this draft of the ruleset was posted. Final rules will have the hexapod league in them, basically, you follow the same rules except you are allowed 6 legs. As for target boards, you'll be following the Quadrapod rules (4 target plates, one must be visible from any horizontal angle, etc).



- In sec 1, are the Safety and Technical Inspection regs listed anywhere for what we have to follow to be allowed to compete? Not sure if this is meant to be from the safety threads here or something specific to RoboGames.

- On the arena walls, are they really mesh? As in a screen with holes? Is it possible to list the material and/or provide samples to participants (cost or otherwise) for safety testing?
This is the entire ruleset, the inspection is mainly intended to be sure you comply with these rules. If you are using either a tank gun or a defender, your inspection will likely consist of us making sure that your target plates work. Obviously, we'll be concentrating more on questionable weapons. As you build your mech, you might consider having a thread here on the boards, or dropping either Andrew (Tyberius) or I a question if a particular item is questionable. Frankly, if you're worried it's outside our rules, I'd suggest avoiding such a design.

Yes, the walls are a mesh, a very fine mesh. BB's cannot escape it. Tyberius may have the part number, I'm not sure. Again, if you are worried that your weapon will pierce the fabric, talk with us before hand about what it is.



- Sec 4.3, is the 3x3 no-hit area tied into the scoring system somehow? It might be easier to just guarantee that the bots will not have visual contact at the start of the match, either though buildings blocking the view or by having them face away from the center?

Mechs will always have cover between them at start time. This area is a 3FOOT x 3 foot area, and is basically our "no german pillboxes" rule. In essence, before you can score on an opponent, you have to prove that your mech is capable of walking while fully loaded. This is not handled automatically by the scoring system, if a mech manages to fire on an opponent before it has left the box, the judges will reset the opposing mech's HP counter.


- On the target system, is that $100 for a 4-board system w/ transponder? I'll try to dig up the current thread for targets, but it's 3" x 3" x what? Are there any requirements on the placement of the transponder?
Final details for next year are still being worked out. Each target plate will likely be $10-12, the transponder should not be more than $50. They were heavily subsidized this year, so don't read to much into the very low prices of the 2009 boards. The transponder can go anywhere in the bot, only the plates are governed.

-Fergs

tom_chang79
08-21-2009, 03:45 AM
There was an idea thrown around at Matt Denton's forum (hexapodrobot.com), which is nothing new, but always comes up, why not incorporate laser tag instead of live rounds?

Those "laser tag" toys they sold in the 80s were nothing more then IR emitting diode and a photosensitive diode?

The only thing I don't like about using the light-based stuff is that there's not much limitation on distance. You can score a hit across the the room sort of speak... But with nerf rounds, gravity will always pull it down (so ranging is important)...

robokoi
08-21-2009, 06:21 AM
There was an idea thrown around at Matt Denton's forum (hexapodrobot.com), which is nothing new, but always comes up, why not incorporate laser tag instead of live rounds?

Those "laser tag" toys they sold in the 80s were nothing more then IR emitting diode and a photosensitive diode?

The only thing I don't like about using the light-based stuff is that there's not much limitation on distance. You can score a hit across the the room sort of speak... But with nerf rounds, gravity will always pull it down (so ranging is important)...

I think the point of live rounds is that it is also something the audience can see. Like battlebots, it's good to see physical interaction because that is harder. It's not pointing something and going "bang, bang. you're dead".

It's also a harder engineering problem to solve. In addition to gravity, there's load issues, limited ammo, and best of all.... multiple weapons choices! Airsoft, Nerf, melee.... can you think of a secondary weapon or a second class (BB, microrockets) that could participate in a lasertag competition?

DresnerRobotics
08-22-2009, 12:13 AM
There was an idea thrown around at Matt Denton's forum (hexapodrobot.com), which is nothing new, but always comes up, why not incorporate laser tag instead of live rounds?

Those "laser tag" toys they sold in the 80s were nothing more then IR emitting diode and a photosensitive diode?

The only thing I don't like about using the light-based stuff is that there's not much limitation on distance. You can score a hit across the the room sort of speak... But with nerf rounds, gravity will always pull it down (so ranging is important)...

It has been brought up before and frankly, I find laser tag a complete bore. It's one of the things that I knew from the beginning wouldn't be an option in my mind. There is something more exciting about having actual ballistics firing, for both the crowd and the competitors. You can hear the airsoft guns firing, see the BBs flying, and hear the ricochets as they make contact; far more exciting than invisible, silent, lasers.

Darkwolf3128
09-16-2009, 12:05 AM
Is the does the hardcore weapon set allow for metal or plastic BBs? I wasn't too clear on the meaning there.

Adrenalynn
09-16-2009, 03:17 AM
Yes, it does.

darrellt
09-16-2009, 05:49 PM
Hi guys,

A buddy and I are currently working on a brat based mech. He walks and shoot thanks to the excellent tutorial by the lynxx guys. We are prototyping an Android phone as the camera/brain.

I have two questions regarding rules for autonomous bots:

1. Does the remote cut off rule require a separate radio/micro-controller system? Can the cutoff in the software of our primary control system? Does the scoring system have an extra output for cutoff use?

2. An autonomous bot will likely require an bright specific color on the targets to have any hope of finding the other bot. If we provide day-glow post-it notes, will you guys be ok with sticking them to your target panels? It will not require any permanent modification.

lnxfergy
09-16-2009, 06:08 PM
1. Does the remote cut off rule require a separate radio/micro-controller system? Can the cutoff in the software of our primary control system? Does the scoring system have an extra output for cutoff use?

You will have to add some sort of radio system to shut it down. We're not necessarily asking for a redundant system, so you could use your main processor for this, as long as you can verify it works. In the airsoft division it's not a terrible problem (heck, squidword went awry on us last year, so it's not like the controlled bots are gonna be any better).

The scoring system WILL NOT handle cutoff. The scoring system does one thing: handles scoring. We've had numerous requests over the past year and a half for features to be added to the scoring system, and I'd just like to reiterate this fact (not jumping on you darrell, just want this note here for any future readers). Due to the issues of XBEE radios, communication is actually uni-directional, a scoring board never hears back from the central unit.


2. An autonomous bot will likely require an bright specific color on the targets to have any hope of finding the other bot. If we provide day-glow post-it notes, will you guys be ok with sticking them to your target panels? It will not require any permanent modification.

I don't see any issue with this really, I know I'd be fine with it. However, I would seriously suggest havinig a non-autonomous mode in addition to anything you try to do autonomous. Take a look at some of the pictures from RoboGames, the lighting is wretched, and we really can't do much about this. You might also search the forum, there have been several threads about autonomy and mechs which may provide furthur insight.

-Fergs

Adrenalynn
09-16-2009, 07:33 PM
If we have a real demonstrable honest-to-goodness autonomous 'bot demonstrated (video is fine) working before the event, I will commit to evenly lighting the arena, and will provide [approximate] expected temperatures in advance. I have the lighting gear to do it, but I don't want to commit to it (a *LOT* of work) if there's not really going to be a contender.

darrellt
09-16-2009, 07:34 PM
Ok thanks. The android phone has 802.11b so that is how i will connect the software kill switch and view the video.

We will defiantly have non-autonomous mode also since that is a requirement before we can even dream of autonomous. The reason I think autonomous has become possible is that the phone has camera/compass/accelerometer/wifi/bt/gps/serial port/battery/500mhz processor all in one small unit that can be programed in java. Since we built from a tutorial, a lot of the mechanical stuff was already figured out for us, so we have more time to program.

Adrenalynn
09-16-2009, 07:35 PM
We posted simultaneously. See my reply [aka BUMP!]

oneday
09-30-2009, 02:24 PM
Hello, a newbie here. I was reading the rules and I don't know if this was ever discussed, but I would like to put forth the following:

This might be a future enhancement as the robots become more sophisticated.

As this is to emulate MechWarrior and the others, I thought that hit induced damage could be applied to the mechs. As hits are accumulated, use a random assignment to disable some functionality of the mech, ie., after so many hits the left gun is unusable or the left leg has 50% less power.

Just a idea.

lnxfergy
09-30-2009, 02:31 PM
Hello, a newbie here. I was reading the rules and I don't know if this was ever discussed, but I would like to put forth the following:

This might be a future enhancement as the robots become more sophisticated.

As this is to emulate MechWarrior and the others, I thought that hit induced damage could be applied to the mechs. As hits are accumulated, use a random assignment to disable some functionality of the mech, ie., after so many hits the left gun is unusable or the left leg has 50% less power.

Just a idea.

Such things have been discussed before. The issues are:


Given current technology, and how close these bots are to fully loaded, 50% power loss in a leg == bot dead
Many mechs only have 1 gun
It would also likely be left to the end user to implement (we can't possibly expect the scoring system to be inserted into the middle of every robot and be that integrated to cause the shutdowns itself, so ultimately the shutdown procedure is left to the user).

-Fergs

DresnerRobotics
09-30-2009, 02:36 PM
Yeah I love the idea in concept, the implementation of it currently is just far beyond our reach. Definitely something to aspire to though!

sam
09-30-2009, 05:46 PM
Oh!

I have a question that I have been thinking of.

I know that in the Mech games, the mechs often have the ability to fly, so I was wondering if it would be legal to make a quad (or whatever) that could lift off the ground? I know the cage isn't that big, so the flying would have to be well controlled. The quad would also need to have a proper walking gate of course.

I also know that you don't want to break the opponents mech with the propellers, but something this (http://www.hobbycity.com/hobbycity/store/uh_viewItem.asp?idProduct=6630&Product_Name=EDF55_with_C20_Motor_Assembled__203W) could be used. So the props couldn't hit the other Mech.

Just an idea I was fondling.

Sam

mannyr7
09-30-2009, 05:51 PM
I have an Eject button on my HUD to trigger a death sequence in the unlikely destruction of my mech, just for haha's. ;) The audience should love it though. That's as far as I wanna take it right now, as Ferg's said these things are straining the limits of affordable servo technology.

Adrenalynn
09-30-2009, 05:51 PM
You come up with a mech that can walk carrying all that weight, shoot, be remotely piloted, and do so well enough to be competitive under the rules as-writte AND make it fly and it's not gonna be a problem getting it in - promise...

Don't hold your breath, though... Blue isn't a particularly good color on you. ;)

sam
09-30-2009, 05:55 PM
You come up with a mech that can walk carrying all that weight, shoot, be remotely piloted, and do so well enough to be competitive under the rules as-writte AND make it fly and it's not gonna be a problem getting it in - promise...

Don't hold your breath, though... Blue isn't a particularly good color on you. ;)

Hum...

I was wondering if I should write "hypothetical question":) but you replied before.

But trust me, I will try, if it'd be ok for the competition.

Sam.

DresnerRobotics
09-30-2009, 06:28 PM
It would be fine, but you'd likely find DARPA knocking on your door offering you a job very quickly.

SystemDefect
10-06-2009, 08:30 AM
I would also like to add, I know it would fall into the Gimmick Clause, but I feel it should be addressed in the final set of rules. The Monitoring of other players cameras, interference or jamming, in any way shouldn't be allowed. This is to include interference in other player's control systems. Also anti robot stuff shouldn't be allowed like spitting glue at your opponent Water ect... Although if you have a robot that can shoot a mini net around someone, that would be cool and should be allowed.

Adrenalynn
10-06-2009, 12:20 PM
There's the unwritten "Adrenalynn will throw you into the Bay if you get stupid" clause that pretty much covers all of that. I'm assuming since it looks like you're stationed in San Diego (thank you for your service!) that you've been up for Fleet Week in The City. You therefore know how undesirable getting thrown in the Bay would be (unless you're a SEAL in which case it's kinda like the proverbial 'Briar Patch' - and I might have just a moderately larger challenge in executing said rule enforcement... ;) )

darrellt
01-27-2010, 02:10 AM
Is there minimum reset time between matches? I need to estimate how quickly I will need to recharge my batteries, and how many extras to bring.

lnxfergy
01-27-2010, 11:02 AM
Is there minimum reset time between matches? I need to estimate how quickly I will need to recharge my batteries, and how many extras to bring.

Darrell,

I can't say for sure how long between matches, however, based on the number of entries we expect, I wouldn't expect to run faster than an average of one match per hour. I've been suggesting that everyone have at least two sets of batteries.

Friday will be testing/exhibition matches only. We'll most likely be running a double elimination tourney over Saturday and Sunday. If we have 12-16 bots, we'd have about 20-30 matches for a double elimination tourney, that's 5-8 hours of game play across 2 days . The winning bot would go 5-8 rounds over those two days. Obviously, in the later rounds on Sunday, we'll be running a bit faster, but I still don't expect more than 4-5 runs over 3-4 hours.

Personally, Issy's batteries take about an hour to charge, given two sets (and charging each battery before arrival), I could run 4 matches in a 2-3 hour span.

-Fergs

Stobs
04-05-2010, 11:42 PM
Question about BB velocities? I re-read the rules on the official mech-warfare site and I just got done reading this thread from start to finish and I don't see them listed - I know they're listed somewhere because I remember reading specific fps values for airsoft & hardcore. A little help please? TIA :)

PS: Nevermind; I overlooked them on the http://robogames.net/rules/mech-wars.php page (http://robogames.net/rules/mech-wars.php), which under the weapons section clearly states that for the airsoft class the maximum allowable BB velocity is 350 fps and for the hardcore class it's 450 fps.

lnxfergy
04-06-2010, 08:57 AM
Question about BB velocities? I re-read the rules on the official mech-warfare site and I just got done reading this thread from start to finish and I don't see them listed - I know they're listed somewhere because I remember reading specific fps values for airsoft & hardcore. A little help please? TIA :)

PS: Nevermind; I overlooked them on the http://robogames.net/rules/mech-wars.php page (http://robogames.net/rules/mech-wars.php), which under the weapons section clearly states that for the airsoft class the maximum allowable BB velocity is 350 fps and for the hardcore class it's 450 fps.

Check the first post of this thread again, I just updated to the final ruleset -- at the moment, that is the definitive rule set (we'll get the mech wars website updated shortly, not sure how long it will take to get the robogames set updated).

There is no hard limit on FPS (since we really can't measure it) -- I'm pretty confident that no off-the-shelf Airsoft gun, that a moderate walker (i.e. AX-12, hitec servos, not an EX-106 based walker) could carry will not pierce the fabric of the arena -- and therefore is fine. If your custom building a gun, you'd want to get some sample fabric (I believe the P/N and source has been posted, and we'll get that added to the rules as a foot note at some point).

-Fergs

Stobs
04-06-2010, 05:26 PM
Thanks for the heads up. I have to say I'm a bit disappointed about the stipulation against interfering with targeting acquisition, and to a degree wireless control. Hopefully moving forward the ban on interfering with an opponent's mechanisms for acquiring targeting solutions would be lifted/amended - granted, if it can be shown that one's own mechanism(s) for initiating counter-acquisition abilities doesn't inflict more than the most trivial of damage (something in the arena of simply wiping off a lens, at worst). I do appreciate the need to keep costs relatively contained, so shielding one's wireless control (technically speaking well beyond my abilities to implement or render harmless), but the "warrior" in me is shouting out 'all's fair' and all that. A tune I'm sure I'd be less enthused about if it were my grand or three worth of brackets, servos and time standing in the middle of the arena! ;) On the flip side of that last though, just from my short time of visiting here and reading about people's projects, I wouldn't be surprised to see the Mech-Warfare competition further subdivided within three to five years into separate piloted and autonomous classes.

Edit: And while I'm at the Pouting Trough, I might as well add a pfffftt about not being able to use a "sticky" grappling device to trip up an opponent - had that one worked-up in my head the first day I found out about Mech-Warfare! lol :)

lnxfergy
04-06-2010, 05:37 PM
Frankly, as for the interference, once you've built a bot, you'll find out it's tough enough keeping them running in that environment without adding any interference. Most people here are not interested in engineering new robust wireless control strategies and hardware.

The focus needs to be on walking and shooting in a controlled fashion - last year, only a few bots could do that -- and I'd wager a guess that out of our 14 entries, not all of them will perform as well as their creators would like.

-Fergs

Stobs
04-06-2010, 07:06 PM
...Most people here are not interested in engineering new robust wireless control strategies and hardware...

That's exactly why I tempered my comments about control interference.

gdubb2
04-06-2010, 07:21 PM
Hi StoborRobonaut,

I have been on both sides of the combat fence so to speak. First starting with a 120 Lb. Full combat bot. After 7 or 8 years of destruction, I just got tired of having to completely rebuild my bot after each event. One had to be willing to put your work in the arena knowing full well it may come out in small pieces.

When Mech Warfare came along, I found my opportunity to compete in a combat setting without the destruction. This is a perfect fit for those of us that want the competition without the destruction. I hope it doesn't go into the other realm, in my mind that would ruin it.

Have fun at RoboGames..
Gary

RobotAtlas
04-06-2010, 07:51 PM
Forgive me if I missed something in this thread, but does it mean no hexapods?
http://robogames.net/rules/mech-wars.php
> Your robot may have up to 4 legs. There is a penalty for using more than 2 legs.

lnxfergy
04-06-2010, 08:22 PM
Forgive me if I missed something in this thread, but does it mean no hexapods?
http://robogames.net/rules/mech-wars.php
> Your robot may have up to 4 legs. There is a penalty for using more than 2 legs.

The rules on the robogames site are from 2009. The active rules for 2010 are in the first post of this thread (and shortly, will be updated on the mech-warfare.com site, not sure when the robogames site will get updated since we don't directly control that (as far as I know))

Hexapod league was an addition for 2010 -- although we didn't actually get anyone to sign up...

-Fergs

RobotAtlas
04-06-2010, 08:38 PM
The first post says practically the same
> Mechs may have up to 4 legs.

Stobs
04-06-2010, 08:49 PM
Believe me, I'm not trying to make waves and I look forward to growing both a hobby and friendships here, so no offense intended - and I certainly don't mean to come across as taking an "ah, this stuff is easy" attitude at all. Having had a smattering of basic programming courses, and gearing up to take a course in Assembly, I realize this stuff if HARD - but very worth it, imho. I also don't believe in taking oneself overly seriously either, so that may be part of it too. Sometimes text can be a pretty dry format, but I thought I had infused some levity with a bit of a tongue-in-cheek tone...guess I need to work on my writing skills! :} I thought I had stressed that I wasn't looking forward to physically damaging each others mechs [..."wiping off a lens, at worst)..."], since this can obviously get to be a pretty expensive hobby in a hurry, but maybe I tucked that away a bit too much. In any event, I hope that I've explained my position and intents sufficiently. I'm looking forward to the event as a spectator this year and having a good time and maybe getting enough things to sink through my thick skull to help out for when I start the build of my first basic bot shortly after the event....and I promise, I'll wait to start designing that phase-shifting, jump-jetting and Gauss-gun equipped Mech for at least another hour or two! :veryhappy:

Edit:

The first post says practically the same
> Mechs may have up to 4 legs.

Hey RobotNV :)
I think Inxfergy is referring you to two items down on that list - Section I-3 instead of Section I-1. Are you planning on building a hexapod? If so please post some pics!

lnxfergy
04-06-2010, 09:39 PM
The first post says practically the same
> Mechs may have up to 4 legs.

Section I - Leagues of Play

There are three leagues of competition:


The Airsoft Classic league is open to all walking robots with 4 or less legs.

The Airsoft Hexapod league is open to all 6-legged robots. It is intended as an easier entry point into Mech Warfare.

The Hardcore league is open to all walking robots with 6 or less legs.

I've fixed the error that was in section 2 in the document, so that it concurs with this....

-Fergs

RobotAtlas
04-06-2010, 10:27 PM
Thanks guys, now my engineering mind can rest. ;)

I wonder why do you think there are no hexapods.
I would think if you lift 2 legs at a time, the hexapod has 4 more to stand on (vs 2 in quadroped).
Wouldn't it mean a hexapod can carry twice as much weight (well, not quite twice after counting extra weight of 2 legs and possibly more batteries)?

Plus with Biolid King spiders you got half way to a RG hexapod (well, maybe 25% of the way).

lnxfergy
04-06-2010, 10:38 PM
Thanks guys, now my engineering mind can rest. ;)

I wonder why do you think there are no hexapods.
I would think if you lift 2 legs at a time, the hexapod has 4 more to stand on (vs 2 in quadroped).
Wouldn't it mean a hexapod can carry twice as much weight (well, not quite twice after counting extra weight of 2 legs and possibly more batteries)?


Yeah, well. Most people attracted to this competition seem to be attracted to the challenge -- so things like a quad or Biped are a natural choice. I'd imagine you'd see more hexapods if a certain someone hadn't touted a certain Biped platform as being a natural choice for beginners.


Plus with Biolid King spiders you got half way to a RG hexapod (well, maybe 25% of the way).

King spider + ArbotiX Starter Kit == Issy on 6 legs (NUKE supports hexapods, so you'd have an IK powered hex) *

-Fergs

* Disclaimer: I'm the creator of the ArbotiX, and the guy who built Issy (which won MW09)

RobotAtlas
04-06-2010, 11:00 PM
> Biped platform as being a natural choice for beginners.
Reading the posts here, it seems bipeds are the most advanced (and expensive ones).
Continuing my logic, a biped stands a lot on one leg while walking, so roughly speaking it needs 2 times as much torque as a quad and a whoping 4 times as much as a hex.

Translated into $:
for hexapod: Robotis AX-12+ (http://www.trossenrobotics.com/dynamixel-ax-12-robot-actuator.aspx) 16.5kgf/cm $50
x 4 of torque for biped:
Robotis RX-64 (http://www.trossenrobotics.com/dynamixel-rx-64-robot-actuator.aspx) 64kgf/cm $300

$300/$50 = 6 times more expensive.

Did I make a mistake somewhere?

Of course you can't deny sexiness of bipeds, but look at the robustness, speed and stability of Issy walk vs. any bipeds.

Please, please tell me if I'm mistaken somewhere.

lnxfergy
04-06-2010, 11:37 PM
Did I make a mistake somewhere?

Of course you can't deny sexiness of bipeds, but look at the robustness, speed and stability of Issy walk vs. any bipeds.

Please, please tell me if I'm mistaken somewhere.

Nope, no mistakes there, you've drawn the logical conclusion.

-Fergs

lnxfergy
04-26-2010, 12:53 AM
To All 2010 Competitors:

Please leave your feedback/recommendations here regarding rules evolution. Anything you feel needs to be changed, thoughts on portions that need clarification, etc.

-Fergs

darkback2
04-26-2010, 01:04 PM
First off I want to say thank you to the event organizers, that is everyone who helped to make mech warfare happen. Thanks to Fergs we had a scoring system, and Connor for doing the work with the network. I understand that there were challenges for all of you. Ofcourse thank you Tybes for organizing a fun event that we could all participate in, and Fon for the really cool buildings.

I think we need to reconsile certain aspects of the rules in terms of what we are assessing. What do we want the winning robot to do/be, and how might robots benefit from more/different types of classes.

I had a great time seeing and playing with everyone, but I also fear that at the event, too many wins were either through technicalities, or failures on the part of the loosing robot's network/camera. I would propose switching the way in which the event is scored to reflect the actual performance of the robot.

One idea would be 3, 3 minute rounds judged on the following criteria using a 4 point scale with the exception of hitpoints.
- Total hit points
- Robot mobility
- Agressiveness
- strategy

And a total overall score for the challenges taken on by the builder. Quad vs Biped, bipeds being given an edge in all catagories possibly encompasing a weighted score multiplyer.

Robots would not be moved between rounds, as in where you stop moving is where you start moving, unless the round ended in a shoot out in which case the robots would be seperated by a few squares (think respawn) to give both an opportunity to have another shoot out and to encourage the event to be based on more than who has the most powerful guns. Winning 2 out of three rounds would be the match.

I feel that this would encourage strategy, aggressive behavior, and help eliminate a robot winning based on some technical fault.

I would also suggest we add a pre-qualification requirement (possibly for second year competitors). All builders should have to submit a video prior to the start of the event demonstrating that they can successfully:

- walk (come up with a distance)
- turn
- shoot

Also prequalification could include getting mac addresses for the network to minimize communication errors.

I guess all of the above would require impartial judges which further complicates matters a bit...

Finally...the scoring system and network/controller problems. Failures in the scoring system, or network that are completely out of the control of the participants should not cost the participants to loose matches. The fact that a number of wins were simply due to communication errors (time of day of the match) or scoreing system errors (the system registering the robot firing or walking) seams a bit unfair, especially when we consider that several participants recieved their scoring systems on friday of the event. I propose a "good spirit" clause in which an impartial judge can decide if a problem is the fault of the builder, or nature, or the result of a participant doing what we are encourageing them to do. (gary getting his targeting plate ripped off by ramming up against giger's legs). I would also propose that bracket matches be held before or after the event when the network/radio traffic is at a minimum. Exhibition matches would be held during the event in which we could experiment with different combinations of 1on 1, 2 on 2 and so on.

Please don't take from this post that I did not have a great time at the event, I had a blast. I'm just looking for ways to make things better for everyone in the years to come, to encourage participation,

Best

DB

lnxfergy
04-26-2010, 02:02 PM
I totally agree about network/camera problems -- towards the end of the day we were assessing the possibility of the back pit room (which has an entrance right off the main gate) as being the new home of MW -- which should shield us from some of the RF interference.

I'm not too keen on shorter rounds, because, frankly, very few robots have much mobility -- out of 13 bots, I would guess only 4-5 could actually cross the arena in 12 minutes.

Prequalification is almost certainly going to be on the docket next year: my initial thoughts are that on Friday (anytime before 7PM) you must show that your bot can start in the box, walk past the first building, shoot one of those 12" sticky targets within something like 6 minutes. We'd also have a gun on hand to fire on each target plate and be sure that you have your plates attached correctly. By getting this out of the way on Friday, we'll be ready to seed the bracket late Friday night, and go to matches first thing Saturday -- we'll also be more clear about the actual competition time so that people don't arrive at 11:30, and have a match at 1:00.

RE: scoring panels. We actually had a lengthy discussion about this yesterday towards the late afternoon. I wasn't actually around for either of Gary or Seth's panel-based losses -- but the plan for next year would be that if a plate falls off, and it was using the designated 2-strips of velcro or the 4-bolt pattern, we would pause, reset the panel, assess a single hit (if the panel falling off didn't cause a hit, much like the biped falling over rule). We were also suffering some RF interference later in the afternoon, and manually watching the blue LED to assess hits that the XBEE radios missed (and for that reason, we're going to be more explicit about the mounting of the LED for next year, so that we can recover in the event of network outages). I know early on Saturday we had some issues with people who hadn't brought their panels/transponder up to test it, that sort of thing will go away next year by virtue of the pre-qualifications.

-Fergs

(I'm actually flying back today, so I'm slowly posting thoughts into this thread...)

darkback2
04-26-2010, 02:32 PM
Sorry,

The shorter rounds were not necessarily end points, or stopping points, and the robots would not go back to the beginning point at the end of a round unless they were in s ahootout. This would allow bipeds 12 minutes to cross the arena with short 30 second - 1 minute pauses. It would discourage a quad from crossing the arena scoring a few hits/no hits, and then hiding for the rest of the 12 minute round. Basically I think a lot of the wins were single short burst shootouts. Having 3 rounds would have the following results.

1 or 2 tied rounds (robots walking across the arena, followed by a win

2 wins from robots walking around and having 2 shoot outs and then respawning which gives the other robot a (fighting) chance

2 TKOs from judges deciding which robot had the higher score at the end of each round if neither robot is ever engaged in a shootout.

I'm just trying to encourage participants to:
- be agressive (something we want)
- have a robot that walk (squidword included)
- engage in more than one crowd pleasing shootout per 12 minute match if possible.
while
- discouraging a lot of the problems we had, and making it so that firepower is less of a deciding factor. (simply dumping your gun once a match shouldn't do it.)

DB

Stobs
04-26-2010, 02:40 PM
I'd also like to send out kudos to everyone involved with the event! Unfortunately due to personal obligations I wasn't able to make it to the event Saturday and Sunday as I had planned, but being there for most of Friday was enough to glimpse a bit of how much work is involved in pulling off the event - which is obviously quite a lot. I don't know who does what but those that I spoke to were all very friendly and helpful, and I'm very much looking forward to being a fellow competitor next year rather than simply being a spectator. Nevertheless, even as a setup-day spectator I thought the arena looked great and gained a much deeper appreciation of the work involved in hosting and participating in Mech-Warfare - here's to continued success with the event and the evolution of the participants! :D

Upgrayd
04-26-2010, 10:35 PM
Just my thoughts so far. I'm sure ill add more later.

Judging / Scoring
For the overall integrity of the event under no circumstances should either an arena judge or someone keeping score be a competitor in the competition.

Movement / Turret Mode / Hiding
All these topics are sort of related.

The movement rule is too ambiguous. At the moment a mech is able to take a 1mm step every 20 seconds or simply fire a shot and be considered 'moving'.

This basically encourages going into turret or camping mode. When one mech is camping it forces the opponent to try to score just a few hits and hide the rest of the match leading to a very uninteresting battle.

Pilot Stations
Pilots need to at a minimum have their backs to the arena during a match.

We also need to address communication between competitors and viewers. At numerous times we had questions called out by a competitor (myself included) that was answered by a spectator change the outcome of a match.

Pit Stops
This is probably up for debate but to me it seems awkward to me that when one competitor calls for a pit stop both competitors are allowed to make adjustments.

Stobs
04-26-2010, 11:30 PM
Hi Upgrayd - first off I'd like to say it was very nice meeting you on Friday and I hope you had a good event! :)


Just my thoughts so far. I'm sure ill add more later.

Judging / Scoring
For the overall integrity of the event under no circumstances should either an arena judge or someone keeping score be a competitor in the competition.
...

Do you mean in regards to the active match in particular, or the event overall?

Connor
04-26-2010, 11:35 PM
Just my thoughts so far. I'm sure ill add more later.

Judging / Scoring
For the overall integrity of the event under no circumstances should either an arena judge or someone keeping score be a competitor in the competition.

I can understand where your coming from on this... However, I just don't see this happening any time soon.. I personally provided the wireless networking equipment, the touch screen computer used for scoring, and the two USB LCD 9" panels this year. It takes people to run the matches, at minimum three (1 to man the scoring computer, 1 to watch for unregistered hits, and someone who can run it and setup bipeds if they fall or pull bots apart etc..) and I don't think anyone is going to pay for a flight, hotel, food etc just to come down and run the match and not have a chance to compete. I came this year to redeam myself from last year.. but more importatly, to have FUN this time. Which I did!



Movement / Turret Mode / Hiding
All these topics are sort of related.

The movement rule is too ambiguous. At the moment a mech is able to take a 1mm step every 20 seconds or simply fire a shot and be considered 'moving'.

This basically encourages going into turret or camping mode. When one mech is camping it forces the opponent to try to score just a few hits and hide the rest of the match leading to a very uninteresting battle.

To be frank, Camping, Ambushing and Flanking are all valid tactics, even in the board game. Bipeds have a very hard time, this is only the 2nd year, and we had what, 2 or 3 working bipeds? The bipeds have a several issues, including having to deal with BB's under their feet. Speed, and stability. I don't want to hear, you should build a quad then.. because, again, the vision of Mech Warfare was that of BattleTech and Mech Warrior.. Quads and now Hex's are allowed to increase the number of people who are able to compete.



Pilot Stations
Pilots need to at a minimum have their backs to the arena during a match.

We also need to address communication between competitors and viewers. At numerous times we had questions called out by a competitor (myself included) that was answered by a spectator change the outcome of a match.

Did any of the pilots stand up or look over the canvis part of the arena?

Yes, Communication between competitors (and even team members) and viewers needs to be addressed.. They're was some issues with that.. again, this is all in fun..



Pit Stops
This is probably up for debate but to me it seems awkward to me that when one competitor calls for a pit stop both competitors are allowed to make adjustments.

I'm completly disagree with your take on this. In almost all sports, if one team takes a timeout, the other team does too. The only thing that doesn't is something like NASCAR, which, is a Race..

Now, what I want to see.. is mandatory check in of the mech on Friday. Everyone have their Camera and PORTABLE computer with static IP's. the MAC address and SSID of the Camera registered with the network admin.

Thanks, Connor

Stobs
04-26-2010, 11:52 PM
... Now, what I want to see.. is mandatory check in of the mech on Friday. Everyone have their Camera and PORTABLE computer with static IP's. the MAC address and SSID of the Camera registered with the network admin.

That might be an issue for people who can't get the Friday off or need to travel on that day; what about making a combination check list/data capture sheet? The check list could include yes/no questions that could help designate tentative bracketing and provide a public guideline (loose and/or strict) for performance standards. The data sheet would also be a guideline of sorts, since if the info needs to be captured for the event then the combatants really should be generating it to begin with - yes, that's definitely pointed at myself as I am the type to misplace my glasses on the top of my head! ;)

lnxfergy
04-26-2010, 11:54 PM
Personally, I'm thinking the "pit stops" are just going to go away entirely. The way the rule was written originally (Section V, paragraph 1, "Builders will be give one 2-minute extension to use throughout the weekend.") it was intended to be an extension at the beginning of match -- we of course were extending far beyond this (and will have to avoid doing this in future years), and had agreed on Saturday to allow this to become a 5-minute timeout at any time during a match.

However, a timeout causes some serious issues. For instance, we had a match where one bot called a timeout because they were no longer functioning, and during the timeout, the other bot caught on fire. This is already a double-elimination tourney, and adding the timeout effectively makes it a triple-elimination tournament (since you can save yourself one elimination, possibly, by fixing something on the bot).

If we were to keep this timeout, we'd have to decrease the time for sure -- because if it takes several minutes to power on the bots, a 5-minute timeout and 12-minute match may be up around 25 minutes of runtime. Asking the other competitor to stand there for 5 minutes doesn't work so well.

-Fergs

Connor
04-26-2010, 11:59 PM
That might be an issue for people who can't get the Friday off or need to travel on that day; what about making a combination check list/data capture sheet? The check list could include yes/no questions that could help designate tentative bracketing and provide a public guideline (loose and/or strict) for performance standards. The data sheet would also be a guideline of sorts, since if the info needs to be captured for the event then the combatants really should be generating it to begin with - yes, that's definitely pointed at myself as I am the type to misplace my glasses on the top of my head! ;)

RG is Friday, Saturday and Sunday event.. All COMBOT competitors are required to do safty check in and such on Friday. We have to have people in on Friday to do a systems check with the networking hardware and scoring system. Doing it on Saturday, or before a match just doesn't work for keeping things going.


Thanks, Connor

Stobs
04-27-2010, 12:08 AM
Prequalification is almost certainly going to be on the docket next year: my initial thoughts are that on Friday (anytime before 7PM) you must show that your bot can start in the box, walk past the first building, shoot one of those 12" sticky targets within something like 6 minutes. We'd also have a gun on hand to fire on each target plate and be sure that you have your plates attached correctly. By getting this out of the way on Friday, we'll be ready to seed the bracket late Friday night, and go to matches first thing Saturday -- we'll also be more clear about the actual competition time so that people don't arrive at 11:30, and have a match at 1:00.

whoops, missed this post until after posting to you Connor; if Friday evening/night is going to be a hard-and-fast pre-qualifying then please temper my comments with that in mind.




RE: scoring panels. ... but the plan for next year would be that if a plate falls off, and it was using the designated 2-strips of velcro or the 4-bolt pattern, we would pause, reset the panel, assess a single hit (if the panel falling off didn't cause a hit, much like the biped falling over rule).

Would the hit penalty be applicable even if it fell off due to an issue of manufacture rather than installation (I'm presuming that you're referencing instances where proscribed fasteners loosened up)? If any cause for a plate to fall off is going to come under the heading of 'life happens, here's your hit,' then so be it, but the ruling should be noted either way, imo.

Stobs
04-27-2010, 12:29 AM
RG is Friday, Saturday and Sunday event.. All COMBOT competitors are required to do safty check in and such on Friday. We have to have people in on Friday to do a systems check with the networking hardware and scoring system. Doing it on Saturday, or before a match just doesn't work for keeping things going.

Again missed this one, but took me awhile to write that last post with putting doggie outside :}

sthmck
04-27-2010, 01:58 PM
I am not sure exactly what happened when Gary's scoring plate came off, but the discussion that came about after the second one basically came out to this.

If your target plate which is properly attached to your mech falls off you will be allowed to attach it and will have one point reduced. This assumes that you met the target plate mounting requirements and that you have passed inspection.

On the other hand if the target plate coming off is a result of mechanical failure due to inferior construction you will be disqualified. If you have a pit available then you are permitted to repair your robot therefore avoiding a DQ.

Upgrayd
04-27-2010, 04:24 PM
Reading back on darkback's post I think he hit the nail on the head. We need to figure out what a winning mech is. Are we simply supposed to take the opponents HP down to zero? Are we prepared to have a quad mech loaded up with two cameras and guns pointed 90 degrees apart who just camps in a corner and defends? Can mechs continue to score at least one hit on the opponent and spend the rest of the match hiding to win?

Are we modeling this exclusively around recreating some fond nostalgia of a board game and video games or do we want to develop a competition that is fun for both the audience and the participants?

Is this an engineering and strategy competition or a model and replica building competition that happen to be able to shoot?

muc
04-27-2010, 05:00 PM
Suggestions for problems associated with scoring panels due to network outages

1. A software update and adding another LED. The second LED would illuminate and stay on after a given hit threshold is reached. This would take precedent over the score at the table. Would be easier to monitor if the network goes down. Or use the current LED and flash at whatever frequency once the threshold is met.

2. Maybe this is already being done… transmit the total hit count to the control center on each hit. Also transmit hit count every X seconds even if no hit was made.

3. Drop the baud rate to the lowest possible on the xBee

Connor
04-27-2010, 07:16 PM
Suggestions for problems associated with scoring panels due to network outages

1. A software update and adding another LED. The second LED would illuminate and stay on after a given hit threshold is reached. This would take precedent over the score at the table. Would be easier to monitor if the network goes down. Or use the current LED and flash at whatever frequency once the threshold is met.

2. Maybe this is already being done… transmit the total hit count to the control center on each hit. Also transmit hit count every X seconds even if no hit was made.

3. Drop the baud rate to the lowest possible on the xBee

Part of the problem with that is, we need to be able to assess, reset and or subtract HP's on the fly. We would have to have a two way communication system going between all the Xbee's. I'm not exaclty sure how that works when they're used in the current setup. Fergy?

Thanks, Connor

Connor
04-27-2010, 07:23 PM
Reading back on darkback's post I think he hit the nail on the head. We need to figure out what a winning mech is. Are we simply supposed to take the opponents HP down to zero? Are we prepared to have a quad mech loaded up with two cameras and guns pointed 90 degrees apart who just camps in a corner and defends? Can mechs continue to score at least one hit on the opponent and spend the rest of the match hiding to win?

Are we modeling this exclusively around recreating some fond nostalgia of a board game and video games or do we want to develop a competition that is fun for both the audience and the participants?

Is this an engineering and strategy competition or a model and replica building competition that happen to be able to shoot?

The current setup is a double elimination, who ever has the most HP's at the end of the match wins. Scoring 1 hp hit, and then hiding the whole time from your competitor is kinda BS in my book. Maybe there needs to be a threshold of so many HP's before a winner can be determined.. I.E. Half your HP's must be gone before then end of the match.. If neither mech has enough HP lost. We call a draw.. and deal with it in the bracket somehow. I hope we can have scenario were we have attack/defend. Ambush, and missions etc.. like protect the convoy..

As for modeling what this competition is suppose to be like.. I think it's a mix of of all of that. Recreatingthe board game and video game to such a point that it's fun for both the audience and the partcipants.. I think we can have our cake and eat it too, and hopefully we can have different classes even for weight and biped vs quad etc..

Thanks, Connor

darrellt
04-27-2010, 07:34 PM
First all, many thanks to the organizers for all their hard work. The whole thing ran quite well, overall.

Ideas for next year:

Scoring system: I am going to prototype a modification to the scoring system that uses a bright IR led to send an encoded IR signal. This IR led will be pointed upwards and there will be several detectors in the tent looking downwards. The upwards pointing should prevent excessive flickering in the cameras if they lack IR filters. My plan is to use this to augment the xbee, so that there are two independent scoring systems. I will have it send the the robot name, total hits, timestamp and pin number for every hit event since startup at a random interval < 2 seconds, so if a signal is missed, hits will still count. The timestamp is on the off chance that the finish is very close, a winner can be reliably determined. The random interval is so that the two bots will not repeatedly broadcast at exactly the same time, causing interference.

If their are IR leds on top of the bot, then I can also use some Wii controllers or an IR only camera mounted high above the match to keep track of the positions of the mechs and enforce the 20 second move rule in a repeatable way.


Additonaly I will write a simple webapp that keeps track of the brackets and all the results, times, etc. Of course it will have to twitter the match results, and possibly text competitors a 15 min warning when their match is about to start. This should not be hard as webapps are my day job.


Wireless Interference:

I have two possible solutions to the 2.4ghz interference problem, One is to make the tent conductive, with copper screen or clear conductive plastic film. This will create a Faraday cage, provided we add a roof and keep the door closed. This would have the added benefit of keeping our signals in, so we could turn on all the SSIDS at once, and they would not leak out and we would not have to constantly re-configure the system. Additionally those using bt or xbee would have to be inside the cage, so the tent would need a driver foyer attached. I will try this on the small scale, but this may not be practical due to expense.

The other possible solution is to switch bands, to 5gz This may be easier, but 5ghz cameras are less available, and more expensive.

billyzelsnack
04-27-2010, 08:43 PM
How about going low tech with the targets? Maybe mechanical/spring based that pop off when hit?

Or..

Maybe get rid of the heavy real guns all together and make it an augmented reality setup. Robots could have an AR marker that is perfectly tracked via a camera. You could go nuts with this with virtual stuff like explosions and smoke. This would also open up the possibility of remote internet matches. Even matches between "real" and virtual mechs.

DresnerRobotics
04-27-2010, 10:00 PM
Why are we proposing an overhaul of the scoring system when it worked 98% of the time? I'd say the focus should be on game balance, fairness, and a better solution for the Wifi interference. I'll post more on my thoughts about rule changes tomorrow! Great job everyone!

lnxfergy
04-27-2010, 10:07 PM
I think the things that need to be addressed RE: scoring are that transponders are mounted in such a way that the xbee antenna is not hidden away (I have a feeling that was the issue we saw on Giger), that the receiving XBEE is mounted central to the arena (the far side was where we were seeing some dropout), and that target plates are well mounted.

-Fergs

P.S. I'll remind everyone to leave the ISP header of their transponder board accessible for future firmware updates -- we'll almost certainly have a few for next year (to support 2-on-2 and other interesting exhibition games).

sthmck
04-27-2010, 10:10 PM
Truthfully I think that the scoring system was pretty good. From what I understand it is a major improvement from last year. We have to face the fact that it is going to be pretty hard to get a system that works 100% correctly 100% of the time. I think that the reliability of the system could be improved if there are some changes to how the target panels are mounted. I found that if a target panel is only mounted in one spot and is not mounted very securely, hits are not always registered.

darrellt
04-27-2010, 10:20 PM
Why are we proposing an overhaul of the scoring system when it worked 98% of the time? I'd say the focus should be on game balance, fairness, and a better solution for the Wifi interference. I'll post more on my thoughts about rule changes tomorrow! Great job everyone!

Its not really an overhaul, its mainly the addition of one IR led and some new software to make life easier for the judges. The location tracking the IR allows should increase the fairness and balance. And less reliance on a 2.4ghz xbee reduces noise for the wifi a small amount.

Anyway if I get something cool working I will post a video for consideration. Otherwise the 2010 system was indeed pretty dam good.

billyzelsnack
04-28-2010, 10:50 AM
I think Tybs was responding to my off the wall comments. :)

In staying with the existing spirit.. I would like to see a smaller footprint for the target system. This would include a published scoreboard API for both zigbee and wifi for those that want to implement it themselves. I'm not against a standard sensor requirement.

I'd also like to see CO2 canisters allowed ( if it is not already ). The motorized guns are heavy and bulky. With a canister you can relocate it somewhere more convenient.

DresnerRobotics
04-28-2010, 12:01 PM
CO2 is 100% a no-go in our current arena. There's no way to reliably test it to the point where it would make the safety officials happy.

The scoring system absolutely MUST stay completely standardized. Same hardware, same sensors, same firmware. Introducing any more variables to the mix is just asking for trouble. I doubt the usefulness of an IR redundancy system as it adds unneeded complexity to the system which could in turn just cause more trouble.

The current scoring system worked amazingly well given the conditions. The VERY few problems we had with it can be easily explained: A) The Xbee needs to be mounted in a central position within the arena. B) We need to verify that everyone has their transponder board mounted in an open, unobstructed positition (or at least the antenna). There was a ton of RF noise there which was the culprit for most of our issues, though I honestly only remember a handful of scores that were missed.

Another thing I noticed was people claiming that hits were registered even when the scoring LED didnt go off (this happened to me when I was piloting one match, my LED was mounted in the bottom frame of my camera so I know for a fact it didnt go off). The scoring panels are not perfect, they have a dead spot in the shape of a cross in the center as well as along the border. This is just something we need to accept, consider it a random chance variable and part of the game. We were doing a pretty good job of testing everyone's target panels each time we went in there so we know they were working.

A few other points (I'm sure I'm going to think of more later).

Next year, if you want to use the central Wifi system, we need everyone's camera SSID, MAC Address, static IP, and their laptop LAN static IP address before hand.

Friday will be our check-in and practice day. You will have until the end of friday to get 100% qualified. This will include a full scoring system inspection (we'll be posting more details on refined target panel requirements), Video system check (either you're fully functional with your own, or the provided wifi system), and we will be adding a qualification round for everybody. This will very likely include being able to walk to the center street, turn 90 degrees, and fire upon a 12" diameter target placed in the center of the arena within 5 minutes, with no more than 3 falls. More details on this later.

If you do not check-in and qualify, you will not be competing in the ladder. It's unfair to those who came prepared and ready to fight. You will still however, be able to participate in exhibition matches, which there will be many more of next year given that we wont be struggling with Wifi interference and/or scheduling conflicts with those atrocious 2.4ghz Spectrum Radios.

We have the option of moving the competition into one of the 'pit' rooms in the corner of the event, so we'd have virtually no interference from the outside events. The drawback is that we'd have less crowd traffic. We'll talk more about it later but I think it's a worthy thing to look into.

In terms of camping, we're going to be discussing this further later, I'll probably start a thread dedicated towards what is fair and what constitutes camping, as well as how we will keep track of it. Perhaps the 'Radar' solution that Fergs and I were discussing could be used towards this.

Connor
04-28-2010, 12:59 PM
CO2 is 100% a no-go in our current arena. There's no way to reliably test it to the point where it would make the safety officials happy.

The scoring system absolutely MUST stay completely standardized. Same hardware, same sensors, same firmware. Introducing any more variables to the mix is just asking for trouble. I doubt the usefulness of an IR redundancy system as it adds unneeded complexity to the system which could in turn just cause more trouble.

The current scoring system worked amazingly well given the conditions. The VERY few problems we had with it can be easily explained: A) The Xbee needs to be mounted in a central position within the arena. B) We need to verify that everyone has their transponder board mounted in an open, unobstructed positition (or at least the antenna). There was a ton of RF noise there which was the culprit for most of our issues, though I honestly only remember a handful of scores that were missed.

Another thing I noticed was people claiming that hits were registered even when the scoring LED didnt go off (this happened to me when I was piloting one match, my LED was mounted in the bottom frame of my camera so I know for a fact it didnt go off). The scoring panels are not perfect, they have a dead spot in the shape of a cross in the center as well as along the border. This is just something we need to accept, consider it a random chance variable and part of the game. We were doing a pretty good job of testing everyone's target panels each time we went in there so we know they were working.

A few other points (I'm sure I'm going to think of more later).

Next year, if you want to use the central Wifi system, we need everyone's camera SSID, MAC Address, static IP, and their laptop LAN static IP address before hand.

Friday will be our check-in and practice day. You will have until the end of friday to get 100% qualified. This will include a full scoring system inspection (we'll be posting more details on refined target panel requirements), Video system check (either you're fully functional with your own, or the provided wifi system), and we will be adding a qualification round for everybody. This will very likely include being able to walk to the center street, turn 90 degrees, and fire upon a 12" diameter target placed in the center of the arena within 5 minutes, with no more than 3 falls. More details on this later.

If you do not check-in and qualify, you will not be competing in the ladder. It's unfair to those who came prepared and ready to fight. You will still however, be able to participate in exhibition matches, which there will be many more of next year given that we wont be struggling with Wifi interference and/or scheduling conflicts with those atrocious 2.4ghz Spectrum Radios.

We have the option of moving the competition into one of the 'pit' rooms in the corner of the event, so we'd have virtually no interference from the outside events. The drawback is that we'd have less crowd traffic. We'll talk more about it later but I think it's a worthy thing to look into.

In terms of camping, we're going to be discussing this further later, I'll probably start a thread dedicated towards what is fair and what constitutes camping, as well as how we will keep track of it. Perhaps the 'Radar' solution that Fergs and I were discussing could be used towards this.

I would also like to point out, that NO ONE successfully competed with their own video system during the match at peak RF interference, with exception to manny, who used two iphones in 3G mode (maybe Miss Alignment, I don't remeber how theirs worked). Ad-Hoc, and Routers using Omni Antenna's just didn't work.. Frame Rate were very low and it was just a pain. I did find that switching from 802.11g to 802.11b improved everything. The 802.11b uses a large band, and is more forgiving in RF noisy environments.

My Access point has the ability to have "virtual" Access points. Next year, I will have a Virtual AP setup for each competors SSID. (on both WLAN's). We will turn all of them off except for those who are competing at the time. We will also use MAC addresses to limit who can connect to the access point at any given time, and maybe even hide the SSID's (I think the TrendNet Cameras can do that) Also, a Circular polarity Antenna mounted above the arena might provide better noise rejection as well.

On the scoring system. I think using a Xbee with a Rubber Ducky, or even a patch panel would all but eliminate the interferance issues as well, again with exception to hitting the dead spots on the targets, which simply can't be help.

darrellt
04-28-2010, 02:56 PM
Ok so IR is out then. Does the firmware only send messages once per hit? If so then I volunteer to write a higher reliability protocol with no hardware changes necessary. The xbee would send its timestamped "hit list" once a second along with its current timestamp and that way if one transmission did not go through then the next would. The scoring computer would only add up the hits that are not during a pause and are after the start of the match. A 16 bit timestamp at .1s resolution gives an hour of recording time. Even with 60 hits the total message would stay under 100 bytes.

This detects power problems, because the timestamp will reset, it also detects wireless problems, because you will get < 1 msg per second.

Easy Peasy, no hardware changes, just a simple reflash.

DresnerRobotics
04-28-2010, 03:36 PM
No offense, because I'm appreciative of the offered help, but Fergs is a PhD level Comp Science major and quite capable of programming the scoring system. The only problems we had with the scoring system was environmental interference/noise. Other than that, it worked flawlessly. This is a system we need to keep as simple as possible, and while we'll be making a few minute changes to it (mostly to update visual output & add 2-way support), we're largely keeping it the same as to not fix what isn't broken.

The fix to any missed scoring data is moving to a less noisy location and moving the Xbee inside the tent.

Point being here that I'd much rather us focus on the larger problems at hand. Wanna develop a low cost 5.8ghz wifi camera system for us? :D

Stobs
04-28-2010, 03:59 PM
...
We have the option of moving the competition into one of the 'pit' rooms in the corner of the event, so we'd have virtually no interference from the outside events. The drawback is that we'd have less crowd traffic. We'll talk more about it later but I think it's a worthy thing to look into.
...

What about having a small area that could serve as a crowd teaser in a traffic area and double as at least the mobility qualifying area? I'd think that the time and expense of setting up a second safe and secure arena is counter productive and premature at this point, but having an open area (so that at the diagonal a mech would be able to meet the 'standing start and turn 90' requirement of the newly up-sized arena), could be advantageous. If there was nearly a doubling of last year's mech count this year, then I would guess that with the favorable press Mech-Warfare's gotten so far there may well be close to a doubling of contestants again next year.

If this concept is worth floating around then I'd also recommend that in order to help maintain a safe environment that mech's in this area would need to be confirmed disarmed/empty loads. As the ability of mechs evolve the secondary area could transition in future years from a flat elevation to that of one with multiple elevations, which would help gauge the layout of the main arena for the next year - as I recall there were at least one or two people who had asked about the mech's being able to climb up on the buildings, and while I now realize that's way easier said than done currently, I'd be surprised if such capability isn't on the horizon.

darkback2
04-28-2010, 04:08 PM
I kind of feel like this is getting off topic, and perhaps we should redirect the conversation a bit...not that I started it or am in charge of it.

1) What did people percieve to be the actual problems at mech warfare 2010?

2) What were the imbalances that people percieved/ would like to see addressed?

3) What should a robot have to be/do in order to compete/win?

4) how might any of the above requirements affect Bipeds, quads, hexapods.

We need to know what the problems are/were before we come up with proposals at fixing them.
Hope this helps.

DB

darrellt
04-28-2010, 06:45 PM
No offense, because I'm appreciative of the offered help, but Fergs is a PhD level Comp Science major and quite capable of programming the scoring system. The only problems we had with the scoring system was environmental interference/noise. Other than that, it worked flawlessly. This is a system we need to keep as simple as possible, and while we'll be making a few minute changes to it (mostly to update visual output & add 2-way support), we're largely keeping it the same as to not fix what isn't broken.

The fix to any missed scoring data is moving to a less noisy location and moving the Xbee inside the tent.

Point being here that I'd much rather us focus on the larger problems at hand. Wanna develop a low cost 5.8ghz wifi camera system for us? :D

Point taken, just want to find a way to contribute, since I am clearly obsessed with all things mech.

As for the 5.8ghz project, I did have a physics phd roomate in collage who was working on tiny microwave frequency doublers, but I seem to recall they required a lot of liquid nitrogen ;)

Ok so just as an experiment, I will build a scale model of the tent out of some -50db Nickel/Copper coated polyester screen (http://www.lessemf.com/fabric.html) , and see if a Faraday cage(tent?) seems workable. To me it seemed like the interference got worse when people crowded around with their iphones, so we may have more problems the more entertaining the battles get, even if we move out of the main room.

Stobs
04-29-2010, 12:33 AM
I'm a little surprised that was cast aside with such disregard, but so be it.

DresnerRobotics
04-29-2010, 07:03 AM
I'm a little surprised that was cast aside with such disregard, but so be it.


What was? The proposal to fix something that wasn't broken? I appreciate any help I can get on this, mech warfare is very time and money consuming for me, but I'd rather the help be focused in areas where we actually need it.

Darrel, sounds good. I'd love to have people get some testing done on a faraday cage. I think that combined with relocation of the event would probably do the trick for our interference issues.

Stobs
04-29-2010, 10:29 AM
What was? The proposal to fix something that wasn't broken? I appreciate any help I can get on this, mech warfare is very time and money consuming for me, but I'd rather the help be focused in areas where we actually need it. ...

You raised a concern of relocating the event to one of the "'pit' rooms," thereby encountering a problem with reduced traffic. I offered a possible solution, which if implemented in some form I would think would need to be addressed by the governing rules of the event. I'm not saying it was a good or bad idea but I don't think it was off-topic, and was worth kicking around. But whatever, if you don't want suggestions from people because they haven't built a robot, rather than on the merits of the comment, then I'll just lurk until next year [EDIT], otherwise I'll just continue to try and offer what I can at this point to help support the Mech-Warfare community.

sthmck
04-29-2010, 11:08 AM
You raised a concern of relocating the event to one of the "'pit' rooms," thereby encountering a problem with reduced traffic. I offered a possible solution, which if implemented in some form I would think would need to be addressed by the governing rules of the event. I'm not saying it was a good or bad idea but I don't think it was off-topic, and was worth kicking around. But whatever, if you don't want suggestions from people because they haven't built a robot, rather than on the merits of the comment, then I'll just lurk until next year [EDIT], otherwise I'll just continue to try and offer what I can at this point to help support the Mech-Warfare community.

Was the solution that you are referencing dealing with the scoring? If so no one is saying that we don't want help in making the event better. I think what Andrew is trying to say is that he doesn't feel that what you proposed necessarily the best option. As previously stated by several competitors there are several solutions that will help to eliminate error without unneeded modifications to the current system.
The only real problem with the scoring system is the dead zones on the target plates themselves, and these just cant be fixed without a 3x3 inch FSR.

The actual problem is interference. By moving to a pit room we can cut that problem down significantly. Yes there will be less people watching the event which sucks, but I would fight even if there was no one there to watch. I think that most of the people that competed would agree with me on this.

As I said before we would love to have people help make this event better, but if Andrew thinks that the scoring system is working well enough, there is really no need to continue discussing improvements.

darkback2
04-29-2010, 11:34 AM
I think this has turned into two different discussions. StoborRobonaught was refuring to my trying to redirect the thread after his post about setting up two locations to compete. The idea was, creating a viewer friendly area in addition to competing in the pit area. I think its an idea that should atleast be considered.

I think we should break this thread up into a few different threads.

- Dealing with network/communication issues.
- Dealing with the scoring system
- Dealing with changes to the rules

Just a thought, and sorry for all of the misunderstandings.

DB

lnxfergy
04-29-2010, 11:35 AM
I really wouldn't worry about the crowds.

Let's face it: we had a huge crowd watching even when the bots weren't working and there's been a huge amount of press and web interest. If we move to the side room, that room will be open to the public (our pits would be against the walls, but we'd have plenty of space to bring as large or larger of a crowd than we had). We'd have plenty of signs, they'll find us -- if you build, they will come. Also -- if we have less RF interference, that means matches will be much better -- which draws an even bigger crowd.

EDIT: DB is right, that a separate thread on networking is probably in order. Let's keep it to the rules in here (which I do think includes the scoring system and venue issues).

-Fergs

DresnerRobotics
04-29-2010, 01:52 PM
You raised a concern of relocating the event to one of the "'pit' rooms," thereby encountering a problem with reduced traffic. I offered a possible solution, which if implemented in some form I would think would need to be addressed by the governing rules of the event. I'm not saying it was a good or bad idea but I don't think it was off-topic, and was worth kicking around. But whatever, if you don't want suggestions from people because they haven't built a robot, rather than on the merits of the comment, then I'll just lurk until next year [EDIT], otherwise I'll just continue to try and offer what I can at this point to help support the Mech-Warfare community.

Actually, I had entirely missed that post of yours. I thought your post was in regards to the proposed increased-complexity scoring system, as you posted your response right after I was commenting on that. I always appreciate good/constructive feedback, regardless of whether or not you've built a robot yet.

On the subject of your proposal, it'd be pretty difficult to sell Dave on more space for a 'partial arena' or separate test area. I think lots of signage as well as PA announcements would probably be easier and more effective. Possibly getting Fon to setup his booth outside the room and direct people in, etc.

Stobs
04-29-2010, 02:33 PM
@DB: No worries, you were right in splitting the thread.

@Ty: Ok, I wasn't sure how the space allotment(s) go so that's fair enough.

Brodee
05-23-2010, 07:25 PM
is there a limit to the number of pilots (like pilot + gunner ok?) does this mean each one uses his/her own camera?
can you have many guns?

darkback2
05-23-2010, 07:46 PM
Hey Brodee...

Not sure what exactly it is you are suggesting, so I'll guess and see if I'm right.

Running one mech sort of the same way the military runs a tank, so you have a person manning the guns, and a crew to load it, and a seperate individual driving? I hope this is right. I don't think there should be a limit on this, though it might add a lot to the overall complexity of the project. I'm sure the two or more person team could all crowd around the same monitor, or share the same video feed, but then who would control the camera? In that case you would probably want more than one camera. That way the driver could see where they were going, and the person on guns could see where they were aiming/shooting. Then you would have to build a robot capable of dealing with the extra weight, and the communication issues you would have in getting to a location, aiming and shooting.

Here is what I have found...One person can drive aim and shoot pretty well on their own. Your preffered method of controller being the important part. Play a few rounds of halo. You don't need a seperate person to walk your character around, and another to aim and shoot...

That isn't to say it isn't possible...and who knows, you may just prove that it is preferable. Only time can say.

Good luck, and see you in the arena I hope.

DB

DresnerRobotics
05-23-2010, 08:04 PM
No rules against multiple pilots, how you setup the cameras is up to you. Keep in mind we're only providing each team with 1 AP though, so you'd need to share bandwidth there.

You can put as many guns on your mech as your pocketbook can afford in servos.

Stobs
05-23-2010, 09:19 PM
is there a limit to the number of pilots (like pilot + gunner ok?) does this mean each one uses his/her own camera?
can you have many guns?

man, you're going to beat me to something i've been kicking around for a couple of weeks now, aren't you?! lol I was thinking about finding a hobby partner to learn/build with and then divy up the piloting/navigation/gunnery duties in some manner, with a multi-gunned biped ^.^ (sighs) I'm thinking I won't be able to afford to run through the learning curve (tank, hexapod, quad, basic biped and then KickAbutt!) to get there in just one year - presuming I have the time and the ability to get through that same learning curve that is. Hope it goes well for you and see you in a year! :)

darkback2
05-24-2010, 07:58 AM
Keep in mind we're only providing each team with 1 AP though, so you'd need to share bandwidth there.

In the past people have set up their own networks...not sure if that is still in the books though.

DresnerRobotics
05-24-2010, 08:07 AM
Pretty much everyone who setup their own network this year failed in keeping a solid connection. You're more than welcome to use your own network setup, but if you're using more than one camera it's only fair that you keep your cameras on the same channel as to not hog the 2.4ghz band.

Spartan001
05-24-2010, 11:41 AM
Pertaining to the plastic bb rule, is that for the softcore and hardcore classes or just the softcore?

lnxfergy
05-24-2010, 11:44 AM
Pertaining to the plastic bb rule, is that for the softcore and hardcore classes or just the softcore?

Just the Airsoft league.

-Fergs

Spartan001
05-24-2010, 11:44 AM
So magnetized bb's are in for the hardcore league?

lnxfergy
05-24-2010, 12:47 PM
See 2011 rules here: http://forums.trossenrobotics.com/showthread.php?p=40396#post40396

This thread now closed.