[Official 2006 Game Design] Game Elements and Subtasks

[mechanicalbrain]

I would like to see a came where the drive never has direct line of site (do the entire game through video feed). I would also like to see a game where the space for the robot is limited so flexible frame becomes extremely important.

[ChrisH]

It seems to me that one of the problems with using the cameras was the varying lighting conditions between arenas. Green was not always green and while the calibration numbers might have helped, I think there is a better way.

What about using LEDs (which have a very well defined spectrum) to illuminate a diffusing panel? This would create a significant area of a well defined color that could be detected. The LEDs can be run off batteries and with proper battery pack design they could easily last at least all day without replacement or recharging.

If you really want to get fancy put a polarizing filter on the diffuser and on the camera to reduce the chance of reflections being interpreted as real signals.

I feel this would make "homing" tasks easier by eliminating much potential confusion.

Just a thought

ChrisH

[sanddrag]

For film productions, there is some special stuff they use to do a green screen effect. The backdrop cloth is actually this very special strange type of grey cloth. (sorry I don't know what it is called). Then on the camera is a ring of maybe 50 super bright green LEDs around the lense. The subject (actor) must be placed between the LEDs and the screen, otherwise it doesn't work.

So, due to their color/whatever the LEDs have very little effect on the color of the subject at all but on camera, the grey backdrop cloth comes up a smooth, solid, uniform greeen with no texture. The cloth is made specially to reflect the green light evenly back to the camera. It cannot be seen just by a person, but it is very bright, uniform green on the camera.

So, I was thinking if the robots could have lots of green LEDs and then there could be field elements with this cloth on them. It should make homing in on it very easy with a camera since the color reflected back to the camera doesn't really depend on lighting conditions.

However, I'm not really sure how well this works over distance. I've only tried this operation from maybe 10 feet away.

[robot180]

Back to the tire idea, what if a patform of tires was slightly angled and uneven. This would simulate a more real-life application. The robot would not just drive on perfectly even ground. Some bots could get stuck, but get alliance partners to push them out. Maybe if small objects, like tennis balls, are dropped in the tires, where they cannot be seen by the drivers, the robot could use sensors to find the balls and alert the driver using lights on the robot or controls. When the robot finds a ball, it could somehow reach out of its bottom and grab the ball. Maybe they get points for recovering as many balls as possible, with different colored balls giving different amounts of points. This would resemble a search and rescue situation where the tires are like driving over rubble and the different colored balls represent different types of finds, living people, dead people, animals, and whatever else would be found in a pile of rubble.

[EricH]

For a game object, how about boxes of KK donuts? Human player eats them as the game progresses. (No fair using Dave L. as human player, even during mentor rounds! )

[phrontist]

Wow... tires... that would be great! I love the idea of drastically different size elements too, so I guess the ideal game for me would involve tires filled with ping pong balls. Oh yes...

I'd also like to see some other way for robots to move around, instead of the wheels-on-ground approach. Perhaps aircraft cable(s) running the length of the feild (from driver station to driver station), that robots could grab and winch themselves along on. Essentially, I like the idea of robots off the ground and moving around.

I'm envisioning a feild with a taut aircraft cable across the center (like the midfeild line in soccer) at height x off the ground. You could design your robot to be below height x or you could design your robot to get over the cable using the "zip line". In order to make the zip-line option attractive, the game could revolve around depositing items on the opponents side of the feild at a significant height, like the tetras this year. It would be difficult to deposit items up high if your robot had to fit under that cable, but if you built a mondo arm it could do both functions. Hey! The scroring items could be the aforementioned ping pong balls, deposited in tires. The tires with your ping pong balls could be on the opposing side of the feild, with the goal on your own. You'd have to get across and get back to score, or have on robot get across and send ping pong balls over some how. Robots could do wildly different things and still be competitive!

As an aside, aircraft cable is cheap and strong, and if the feild were designed properly, quite safe (snappage wise). You can anticipate how much tension would be on it in the worst case scenario because you know the weight limits of the robots. You could make them lower for saftey/challenge reasons.

[Billfred]

I think that BEST explored this technique a few years ago, but what some gamepieces with unpredictable attributes?

Suppose this technique were applied to Triple Play. In certain pipes that make up the tetra, some extra mass (nuts? washers? cement?) would be added in. The human players would then have to gauge the weight distribution and put them on the robots accordingly, and drivers would have to fiddle with their technique in the event they picked up a cattywampus tetra from the autoloader.

[Billfred]

Proof that you get inspiration from robots slugging their brains out as well...

Anyone who watches Robot Wars is familiar with the technique to open the pit (a robot, competitor or house, hits a switch made of a tire and a panel, triggering the pit to open). Suppose that a similar device on a FIRST field enabled a device to release a more valuable object.

To apply the example to a prior FIRST game, suppose that beside the ball dump in FIRST Frenzy was a small side chute, containing the mythical 25-point ball. This ball would only be released if a button on the field were held down for a given amount of time (say, ten seconds). Ten seconds is a really significant amount of time in a two-minute match, so teams would have to decide whether going for the more potent ball was worth the time investment. Especially if, in this scenario, their human player was shaky.

[JP_1163]

I see robots walking. I see them stepping over objects of varying heights and widths to reach a goal. I see robots manipulating objects and then returning to their originating spot. I see robots engaged in cooperative ventures where the sum of the parts is greater than the whole. I see robots that don't bash and crash but work together to complete a complex task requiring all partners to work in concert to achieve their goal. I see the scoring based on the number of successful tasks completed within a 2 minute period.......

Wait......wait...... sorry crystal ball is on break......

[anna~marie]

when do we get our clue?

[Billfred]

Originally Posted by anna~marie

when do we get our clue?

Traditionally, the hint goes out a week or so before Kickoff.

[phrontist]

Originally Posted by Billfred

Traditionally, the hint goes out a week or so before Kickoff.

Traditionally, everything Dave says is a hint, but no one has any clue what the game will be.

[Taylor_Ratliff]

For the last 6 months, ive been working at Sam's Club pushing carts with the help of one of those automated cart machines, and let me tell you, THEY SUCK! they break down all the time, water gets inside it, the turning radius is horrible.
The job itsself is absolutely back-breaking, most of the time we were pushing them by hand out in 100F heat (TEXAS!).
I think the idea of an automated cart retriever is excellent, and would make a good theme for a FIRST competition. The robots would have to wrangle scattered carts, line them up, and take them back to the garage area all while avioding obstacles that simulate cars/people.

Another thing is that you cant just push the carts, because if you need to stop they all come apart, you have to have something at the front, so this could easily incorporate teamwork like previous FIRST competitions.

I would also like to say I am against having a human player, as it takes away from the actual design of the robot and encourages teams to go get someone out of athletics for the soul purpose of doing the interaction.

For example, in 2004 our team just went and got a 6'8" guy off of the basketball team to do the shooting for us, and he made 138/141 total shots for the whole weekend. The kid didnt know, or care, anything about robotics he was just there to get a few days off school.

any thoughts, comments?

[mechanicalbrain]

I would like to see the continued use of new sensors every year but combined with the chance to use previous years sensors. Also i wouldn't mind seeing the custom circuit rule modified to allow custom circuits to transmit to the player station. Finally I would like to see the game become more robot intensive. Like, i think the 2004 game would have been much cooler had the robots been allowed to score too.

[dlavery]

Originally Posted by Taylor_Ratliff

For the last 6 months, ive been working at Sam's Club pushing carts with the help of one of those automated cart machines, and let me tell you, THEY SUCK! they break down all the time, water gets inside it, the turning radius is horrible. The job itsself is absolutely back-breaking, most of the time we were pushing them by hand out in 100F heat (TEXAS!). I think the idea of an automated cart retriever is excellent, and would make a good theme for a FIRST competition. The robots would have to wrangle scattered carts, line them up, and take them back to the garage area all while avioding obstacles that simulate cars/people.

Another thing is that you cant just push the carts, because if you need to stop they all come apart, you have to have something at the front, so this could easily incorporate teamwork like previous FIRST competitions.

OK, I promised that there would not be any responses to any of the suggestions submitted to the game design suggestion threads, and no feedback would be provided until the game is unveiled in January. However, I just can't resist with this one.

I just want to point out that this problem has already been solved. Rather than build a robotic shopping cart retriever, just automate the shopping cart itself! The folks over at Derek's Digital have come up with this design:



("Pimp my shopping cart"?!?!)

-dave