You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, our FakeWatchDog implementation will instantly freak out if it gets passed a bad SensorData once.
This might sound good, but it's not useful because it will take time for pressure and temperature readings to transition after a state change, so our WatchDog will probably trip every time we change states.
Implement a storage of past SensorDatas in a new WatchDog class, oldest will be replaced by newest so probably a queue. We will then average/use the median of sensor data readings across all SensorDatas in storage, and use that averaged SensorData to run redlines on.
The text was updated successfully, but these errors were encountered:
Currently, our
FakeWatchDog
implementation will instantly freak out if it gets passed a bad SensorData once.This might sound good, but it's not useful because it will take time for pressure and temperature readings to transition after a state change, so our WatchDog will probably trip every time we change states.
Implement a storage of past SensorDatas in a new WatchDog class, oldest will be replaced by newest so probably a queue. We will then average/use the median of sensor data readings across all SensorDatas in storage, and use that averaged SensorData to run redlines on.
The text was updated successfully, but these errors were encountered: