Team Fusion #364, Bayou Regional, FMS Woes

[Wendy Holladay]

Chuck Kennedy at FIRST would know that

[Natchez]

Everyone,

The executive summary with Fusion is, "All is well with 364 ... and we hope it stays this way." Fortunately and unfortunately, we found nothing wrong with 364's robot during our test runs. Here is the summary of test runs. A "(C)" denotes that the team was receiving camera video at the player station.

Test 1: 364 runs match by themself
Result 1: Normal operations

Test 2: R1-1429(C); R2-bypassed; R3-bypassed; B1-2936(C); B2-364(C); B3-118 (C)
Result 2: Normal operations

Test 3: R1-118(C); R2-1429(C); R3-bypassed; B1-364(C); B2-3847(C); B3-bypassed
Result 3: Normal operations

Test 4: R1-bypass; R2-118(C); R3-bypassed; B1-364(C); B2-3847(C); B3-bypassed
Result 4: 118 dies after being enabled for hybrid mode - Watchdog not fed - similar to what was seen in Connecticut; 364 & 3847 ran without issues

Test 5 (Same as Test 4): R1-bypass; R2-118(C); R3-bypassed; B1-364(C); B2-3847(C); B3-bypassed
Result 5: Normal operations

Test 6: R1-bypassed; R2-bypassed; R3-3103 (C); B1-364(C); B2-118(2xC); B3-418(C)
Result 6: Normal operations

Test 7: R1-2882(C); R2-364(C); R3-bypassed; B1-118(2xC); B2-647(C); B3-2585(C)
Result 7: Normal operations

At this point, we concluded the tests and declared 364's robot good-to-go. They have run a handful of times since the tests without issue; I'll let 364 update everyone with the specifics. The FTA, Jonathan, and his crew were awesome during all of these tests.

Finally, the engineer in me wanted to find a smoking gun with 364's robot but the other part of me is happy to hope that all will be well .... after all, it is the Easter season.

We'll keep everyone updated as the Lone Star Regional progresses,
Lucien

[RyanN]

All is well with us.

We changed one thing. The router. During Bayou, we had tons of problems (obviously), so we used the pit admin spare. Firmware version unknown.

PD Board is still the same. We ran perfectly. I looked through some log files and saw no problems. I'll go through more tonight and report back.

Friday will be the day we would disconnect. Stands were basically empty. No interference.

The difference in the log files I looked are night and day difference from Bayou. We peaked out at 50ms latency at our highest point. Very little packet loss.

We're not holding our breath.

Thanks everyone that has helpe us out so far. We're crossing our fingers. We're here and ready to compete.

I'll keep everyone updated the best I can. I'm stuck with my mobile phone for the weekend.

-Ryan Nazaretian

[Tom Line]

Here's hoping you kick some butt and have a solid regional outting. Good luck.

[JaneYoung]

Originally Posted by b-rant

I just talked to RyanN on the phone. He said the robot is running great and they aren't seeing issues with the Lone Star field.

Ya'll should have seen some of the smiles on the team's faces when the robot was first being tested. It was pretty special to see.

All the best, Fusion.

Jane

[GrimmReaper]

I happened to be in Houston this week working, finished up today. So i plan on stopping by Lone Star to check things out Friday :-)

Good Luck!

[RyanN]

So I browses through the log files and everything looks great.

Trip Time is 5ms on average. We have a consistent spike to 15ms every 30s or so. Non-issue. Assuming it to be a Windows service.

Weird phenomenon though. CPU usage continually increases. Usage starts around 40% and will raise to 45% throughout the match. It's non issue for a 2:30 match, but when practicing for 20 minutes, we can increase to 70% or so.

I wish I could post pictures, but that's a limit to the iPhone.

[rachelholladay]

To Ryan and all of 364: GOOD JOB. I hope its stays good all weekend.
We (Mum and I) were thinking about you guys all day.

[RyanN]

Originally Posted by rachelholladay

To Ryan and all of 364: GOOD JOB. I hope its stays good all weekend.
We (Mum and I) were thinking about you guys all day.

I know. Won't me like 3 Facebook messages. I told her I have text messaging.

[jspatz1]

Originally Posted by Kris Verdeyen

Does anyone know where the field in St. Louis this week was last week?

I was told it was from Bayou.

[Jared Russell]

Originally Posted by RyanN

Weird phenomenon though. CPU usage continually increases. Usage starts around 40% and will raise to 45% throughout the match. It's non issue for a 2:30 match, but when practicing for 20 minutes, we can increase to 70% or so.

This is interesting. What language are you using? If Java, I wonder if this is the garbage collector running. If C++, I don't have a good explanation (probably something in your teams' code). If Labview, I have no clue but suspect it might be memory management related (no idea how Labview actually does memory management under the hood).

EDIT: I assume you were referring to the cRIO's CPU usage. Is this correct?

[RyanN]

Originally Posted by Jared341

This is interesting. What language are you using? If Java, I wonder if this is the garbage collector running. If C++, I don't have a good explanation (probably something in your teams' code). If Labview, I have no clue but suspect it might be memory management related (no idea how Labview actually does memory management under the hood).

EDIT: I assume you were referring to the cRIO's CPU usage. Is this correct?

It's LabVIEW, and CPU usage is on cRIO. I could understand if I was doing something with arrays and processing them, but I'm not.

We're about to have our first match. Fingers crossed. Everything on the field looks good so far.

[Natchez]

Here is a play by play of Fusion in their first match:
Hooked to field
2 shots made in autonomy
alliance bridge to get balls
1 for 3
0 for 2
Coop with 624
...... 364 was fully operational

[linuxboy]

I watched the webcast to see you guys work!! I'm very happy that you're working, the engineer in me is a little sad because I want to know what the root cause of the problem was, but congrats on the co-op balance, and good luck in your future matches.

- Oliver

[Natchez]

2nd match .... fully operational