Aug 8, 2006

More Details on the 2007 FLL Challenge

Dave Doerr was kind enough to email me a link to some photos of the new FLL Challenge setup... the subject is "Nano" and the photos are very interesting.

Check out the photos here.

If you have any questions for Dave, please post comments to this post and I'll ask Dave to check in on it if he can.

8 comments:

Maniac000 said...

Awsome! I'm loving it.

Golfing Fool said...

"More Details..."? Does that imply that there was an earlier post with preliminary details?

Has there been any guidance on how NXT and RCX are supposed to be able to compete on a level playing field during the 2006 FLL Competition? With the advances in accuracy and new sensors, it would seem that NXT-based robots will have an advantage.

dlevy said...

some info on RCX / NXT

http://www.firstlegoleague.org/default.aspx?pid=23850

DullDad said...

My $.02 on NXT vs RCX in FLL:

NXT teams that rely primarily on odometry for navigation should have an advantage over RCX odometry teams, due to increased rotation sensor resolution and servo motor control.

The 2006 FLL Challenge provides lots of opportunity for skilled teams that use landmark navigation because the mat has lines and other markings that contrast with the background and can be used for line-following and positioning via light sensors. However, the tests I've run comparing the NXT and RCX line-following capability show no advantage in performance for the NXT over the RCX when using NXT-G and the current firmware. I've tried creating an efficient line-following program with NBC, but was able to achieve only about the same performance as with NXT-G. I haven't tried RoboLab 2.9 yet (uses different firmware). FLL allows only NXT-G or RoboLab.

I'd like to hear what others have experienced with NXT line-following performance. Maybe someone knows how to take advantage of the NXT where I don't.

I would expect that less-skilled NXT teams will obtain better scores than less-skilled RCX teams because the NXT performs better in dead-reckoning navigation, but that expert NXT and RCX teams will score about the same, since the NXT and RCX seem to offer about the same landmark-navigation performance.

In the long run - post RCX - I think this is a good thing because we should see less-skilled teams do better score-wise than in the past. I hate to see teams getting scores of 10 or 20 or 30.

jeff said...

If memory serves, RIS does not allow the creation of user defined variables. Using variables, I have created a line follower program for the RCX using NQC. I was not able to create such a program in RIS. I have been able to create such a program in NXT-G. Using the Tribot as my base, I have tested the program, and I believe the NXT line follower performs superior to the RCX. There are two significant reasons: first, as mentioned earlier, the built in rotation sensors; second, the NXT-G sensor calibration feature. The sensor calibration feature greatly simplifies the matter of adjusting for varying mat reflectivity and ambient lighting conditions. Click on my name to follow a link to a NXT-G line follower program.

Danny Diaz said...

This doesn't look official - have these been publicly released? If not, I would really hate to be Dave right now, someone is going to have a cow. It's one thing to be shown a mission mat in a closed-door session, it's another to take pictures and post them online when the missions aren't supposed to be released until this time next month!

Jim Kelly said...

I know some teams have received theirs... so it must be in the open.

Jim

lindazoe said...

The FLL field kits ship in August. They come with a CD showing how to assemble each object. Teams may assemble the field objects and speculate about what the missions might be. In mid-September (Sept. 15 this year), FLL posts on the web instructions on where to place the objects on the field mat, and where to use DuoLock (similar to Velcro) to attach the objects to the mat. The missions are described, including the points earned for each mission. So teams could try some programming for having the robot interact with an object before Sept. 15, but they will not know if they guessed right about the actual mission instructions until after the challenge is announced.

I think the field kit is shipped early to make sure all registered teams get theirs before Sept. 15, and to give teams a chance to check for problems with the kit. Last year, I had a defective mat, and I was able to get a replacement before the challenge was announced.

My school district began classes on Aug. 10, so I have already had 2 FLL meetings with my team. The students have almost finished putting together all the field objects, and are eagerly speculating about how each works. I have one object that does not seem to work the way the students think it should. I know that questions like this will be answered by FLL after the challenge is announced in Sept., but I would really like to check with another team to see if their object acts the same as ours. Can you ask Dave a question for me?

Related Posts Plugin for WordPress, Blogger...