Optimal Sensor with embedded PSMs

[JonKing93]

No description provided.

[JonKing93]

Currently, the optimal sensor accepts proxy estimates directly. This is fine for the os.update and os.evaluate commands, but can potentially cause an oversimplification for the iterative os.run algorithm. Essentially, when running the iterative algorithm, the estimates are updated in each iteration via the Kalman Gain. However, this neglects non-linear PSM behavior. In the ideal scenario, the method would instead:

1. Select an optimal sensor
2. Update PSM input variables via the Kalman Gain
3. Re-run the PSMs, and
4. Assess the next optimal sensor using the updated estimates.

To implement this, the optimal sensor will need to be able to store PSMs and an ensemble of PSM input variables. It will also need to implement some sort of error checking to ensure that PSM outputs are valid -- this might be accomplished via PSM.estimate.