Open DominicOram opened 3 hours ago
I'll note for the record that this is a very nice problem to have
I'll note for the record that this is a very nice problem to have
Yh, it's potentially a case of me getting to deep into trying to optimise but worth noting down as it shouldn't be that hard
The robot needs to occasionally dry/soak so that it cleans off any ice that has built up on the gripper. This operation takes ~95s and the robot can't be used in that time. The current logic for this is that every 90 minutes the robot will internally set a flag to say "after the next load dry" and will then dry on the next load. This was good behaviour when we knew a collection would take >90s. However, when we start skipping bad samples this assumption will not be true any more. e.g. the worst case is:
To improve this we should instead trigger the dry/load on the robot once we know we're going to perform a long operation e.g. after we have done an XRC for a native collection and know that we will have diffraction. However, we will also want to make sure the robot does still do a dry/soak at some point even if we are not doing any native collections. Therefore the logic will look something like:
Acceptance Criteria