Saturday, November 29, 2008

Team update, 11/29

On Friday, Patrick, Danny, Joey and Paul worked on missions and some tournament practice. Patrick and Danny tried very hard to make the transition from Andrew's to Patrick & Trevor's programs reliable, but there was too much uncertainty. It's been a hard day for me thinking about it since I could not see how to keep "pattrev" (Patrick and Trevor's mission) in the competition, and I would like to see all of the missions tried in Batavia.

"Danjoe" (Danny and Joey's) and "mn" (Mike and Nate's) missions rely on Andrew's to clear the table. Mike and Andrew returned from their vacation today and had a marathon with the robots. Andrew figured out a way of re-aligning, allowing pattrev to run reliably, choosing to not grab the computer to do so. He also captured the carbon and brought it to base, then pushing all the balls out to storage at once (they were not making it there reliably before. Michael sort of got the ice core captured but ran out of time. It may have been a low battery but it looks like we may have to ditch the ice core. He did make a mechanical improvements, including supporting the cargo arm without making the robot "active", which would have been breaking the rules.

Danny re-ran Danjoe, with the carbon balls in the way. This was to simulate the absence of running the Andrew program. He did this before knowing Andrew would fix the transition to pattrev. This work was important to help make the decision to get those balls out of the way (always run "Mission 1"). He also tweaked the program slightly to avoid a hangup on the levee tester.

Paul debugged a lot of issues with the robot and fixed some reliability issues in the program. The light sensor was not mounted properly and it need mechnical rebuild for his ramp mechanism as was done on the robot used for previous testing. He also reworked the end of his program and reliably makes it to the research area. Still a minor bug to test but it looks good.

Ballpark timing, with no transition time (for changing implements):
Mission Group max pts time
Group 1 (Andrew + Pattrev + Carbon) 110 1:10
Group 2 (Danjoe) 50 0:09
Group 3 (MN) 65 0:51 (less if no ice core for 30 pts.)
Group 4 (fludpt) 70 0:31 (0 or 1 pointer - tbd for 2 or 3)
Total: 295 2:41

From last year we know that transitions can take a significant amount of the time, especially since they are less practiced this year. Realistically, we can run no more than 3 missions. With that in mind, it looks like our best core will be around 200 of the possible 400 points. We will discuss all of this at the next meeting.

No comments: