[INSTRM-1941] moveToPfsDesign failed with maskFile Created: 24/Apr/23  Updated: 25/Apr/23  Resolved: 25/Apr/23

Status: Done
Project: Instrument control development
Component/s: ics_fpsActor
Affects Version/s: None
Fix Version/s: None

Type: Bug Priority: Normal
Reporter: arnaud.lefur Assignee: chyan
Resolution: Done Votes: 0
Labels: EngRun
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Reviewers: chyan

 Description   

tried to use moveToPfsDesign with maskFile but it failed.

(rubin3_ics) arnaud@alefur-pcp:/data$ oneCmd.py fps moveToPfsDesign designId=7362534226687433961 iteration=12 tolerance=0.01 goHome maskFile=/data/fps/maskFiles/SM13_moveAll.csv
2023-04-24T08:55:37.431 sent fps moveToPfsDesign designId=7362534226687433961 iteration=12 tolerance=0.01 goHome maskFile=/data/fps/maskFiles/SM13_moveAll.csv (from arnaud_alefur_pcp)
2023-04-24T08:55:41.643 fps i text="Setting moveToPfsDesign expTime=None"
2023-04-24T08:55:41.643 fps i text="Running moveToPfsDesign with tolerance=0.01 iteration=12 "
2023-04-24T08:55:41.644 fps i text="moveToPfsDesign with twoSteps=True goHome=True"
2023-04-24T08:55:41.745 fps i pfsconfig=0x662cf9deec5c1ce9,86505,Preparing
2023-04-24T08:55:41.794 fps i text="Setting good cobra index"
2023-04-24T08:55:41.843 fps i text="Handling the cobra target table."
2023-04-24T08:55:42.492 fps i text="Reset the current angles for cobra arms."
2023-04-24T08:55:42.492 fps i text="Setting ThetaAngle = Home and phiAngle = 0."
2023-04-24T08:55:42.574 fps f text="command failed: IndexError('boolean index did not match indexed array along dimension 0; dimension is 1181 but corresponding boolean dimension is 2356') in makeTargetTable() at /home/arnaud/devel/ics/pfi/ics_fpsActor/python/ics/fpsActor/najaVenator.py:207"


 Comments   
Comment by arnaud.lefur [ 24/Apr/23 ]

chyan can you please look at my branch.
it fixes the bug, but I want to make sure it's what we want.

It will put the pfi_target_(x|y)_mm to the home position for the masked cobra.
It's a lie but as long as they don't actually move that's okay.

Comment by chyan [ 24/Apr/23 ]

Well, there are two things, assigning cobra targets and move cobra. In theory, we need to ALWAYS assign one target for each fiber. But FPS control which fiber to move based on the index we passed to moveThetaPhi. So, as a workaround we did in last few runs, for ANY masked cobra, we assign them to a location very close to center (0.5+0.5j) and DO NOT moving them.

Comment by chyan [ 25/Apr/23 ]

Tested and close

Generated at Sat Feb 10 16:41:15 JST 2024 using Jira 8.3.4#803005-sha1:1f96e09b3c60279a408a2ae47be3c745f571388b.