PDA

View Full Version : [Official 2008 Game Design] New Technology Discussions


dlavery
05-15-2008, 11:51 AM
This thread is a spin-off of this discussion (http://www.chiefdelphi.com/forums/showthread.php?p=748347), and has been started to present ideas for utilizing the new capabilities that are offered by the new control system that will be introduced in 2009. If there is a control system capability that you always wanted to see highlighted in the game – different development environments, more processing power, alternative communications schemes, new sensor compatibility, enhanced I/O options, etc. and how they can be utilized - this is the place to discuss how those capabilities could be used in the 2009 game. Likewise, while autonomy need not be a part of a specific game, creative uses of autonomous components in any game are sought. Ideas about new drive technologies or inter-robot communications may be reviewed.

-dave

JesseK
05-15-2008, 12:38 PM
Allow teams to bring a [computer] (ITX form-factor with a NVIDIA GTX 8xxx card in it) to the field. The CPU will perform all external interfacing and memory management while the GPU will perform all FOP's necessary for true high-resolution pattern recognition algorithms via the vision system. The libraries for GPGPU are now mature enough for us to be able to do this. Of course, the overall computer would account for ~$1k of the cost of the robot so there may or may not be an advantage for a team to do this.

Maybe I have misinterpreted what I remember from the Atlanta presentation regarding the intended external interfacing on the driver's station. Even if so, further development of the vision system capabilities and usage is very desired.

Chris_Elston
05-15-2008, 02:50 PM
Game Pieces:
I only have one suggestion that I can think of that ties the new controller to a practical "real world" situation. Often times industrial application of the new NI controller offer some type of vision interpolation, I.E. vision guided assembles. Meaning if a vision system is bolted on a six-axis industrial robot arm, it takes pictures of pieces or parts and finds features on the part that will change the interpolation of the command to "assemble" the part or piece to a mate piece. In other words, the vision system on NI should have the abilities to sent x, y, and theta features of a part that can be feed into a coordinate cartesian system on our robots that will alter it's game play or give the robots the ability of detail "sight". Of course this is similar to the vision light, but now you have a greater resolution offered by the NI vision hardware.

Game Sensors:
I'd like to see some implementation of RFID. Again defaulting back to industrial applications, most all high speed packout conveyor systems or warehouse automation pallet systems now have some sort of RFID tracking system. "radio" pucks if you will that can be programmed or just read only. It would interesting to have RFID pucks on game pieces that can be re-programmed by alliance robots. I don't know what the application of the game element would be but say you had a game piece worth 10 points and a robot picks up the game pieces reprograms the RFID tag on the game pieces to change it's point value or the way the game pieces is used. 10 game pieces on the field, and they have to be programmed for "BLUE" or “RED” and everyone scores a common goal, the goal has an RFID reader, that will read what the RFID tags are programmed. If you programmed it correctly to your alliance color you get the score.

Joe G.
05-15-2008, 03:12 PM
Please do not get overzealous with the new system. As great as its new capabilities are, there are still many teams out there who have little to no programming resources. Things like virtually required robot-robot communication would kill off these teams. The abilities can be used, but don't have super-advanced programming as a requirement for success, for those teams like mine that exist largely as an extension of a high school shop class.

rfolea
05-15-2008, 04:47 PM
Game Sensors:
I'd like to see some implementation of RFID... .[/QUOTE]

I second the RFID and would also add something to do with Proximity sensors - they are WIDELY used in industry, are easy to come by and inexpensive.

dlavery
05-15-2008, 04:58 PM
Please do not get overzealous with the new system. As great as its new capabilities are, there are still many teams out there who have little to no programming resources. Things like virtually required robot-robot communication would kill off these teams. The abilities can be used, but don't have super-advanced programming as a requirement for success, for those teams like mine that exist largely as an extension of a high school shop class.

Actually, please don't put any limits on your ideas or creativity at this point. Brainstorm, innovate, and contemplate whatever ideas you might like to consider. We are very aware that not all teams will be able to contend with the implicit expertise requirements contained in many of the ideas. But for now, we don't want to stifle the thought exercise. Reality filters will be put in place later in the process.

-dave



.

Cory
05-15-2008, 05:38 PM
My suggestion:

For 2009, don't do anything that relies on functions of the new controller--since they probably won't work right, out of the box.

Billfred
05-15-2008, 06:58 PM
My suggestion:

For 2009, don't do anything that relies on functions of the new controller--since they probably won't work right, out of the box.Echoed after recalling the bugs of years past. Edit: On second thought, I'll qualify my echo: Don't do anything that relies on the new controller's functions and can't be rigged up easily to work through more primitive means.

That said, suppose that teams could park a robot in a loading zone that loaded red and blue balls from a feeder (think Half-Pipe Hustle) in random order. (The refs would have this knowledge for sanity's sake.) Slightly lower in the loading zone would be a rejection bin for balls of the other alliance's color; teams would receive some benefit from depositing the rejects correctly and leaving with their own load.

I did enjoy seeing a drive-forward autonomous mode be successful this year. From what I've seen, years where that is a viable option (see also: Aim High) have been pretty good years for autonomous. Whatever the game is, I'd be happy to see that be a choice (though not necessarily the best choice).

Lastly, if you want to talk incentive for autonomous: Suppose the winner of the autonomous period (by whatever definition you prefer) got a five- or ten-second period at the end of the match to themselves (read: where the other alliance was disabled). Every single season of the autonomous era, such a boost would be a very interesting change in how the endgame was played.

Chris_Elston
05-28-2008, 08:13 AM
Game Sensors:
Another new sensor that seems to be pretty popular.

Laser ranger finders. (http://www.acroname.com/robotics/info/articles/laser/laser.html)

DARPA uses these all the time. SICK brand seems to be the most popular. However with the rules of the part cost, this sensor cost more than the current 2008 rules allowed. However that was 2008, and this is 2009. :-)

jtdowney
05-28-2008, 08:34 AM
Game Sensors:
DARPA uses these all the time. SICK brand seems to be the most popular. However with the rules of the part cost, this sensor cost more than the current 2008 rules allowed. However that was 2008, and this is 2009. :-)

Laser range finders are definitely a cool idea but don't have to cost a lot. I could see a rookie team doing something like http://www.pages.drexel.edu/~twd25/webcam_laser_ranger.html and besting veterans right out of the gate.