Mean Squared Error

The Mean Squared Error (MSE) is a quantity that can be included in a plot. The MSE is a single number that indicates how closely two other plot quantities are to each other during the entire plot. The closer the quantities are, the smaller the MSE will be.

The Mean Squared Error is typically used during manual tuning as a measure of how close the Actual Position, Velocity, Pressure or Force is tracking the Target. As the tuning progresses, the MSE should become smaller, indicating that the actual value is tracking the target more closely. The actual value of the MSE is not necessarily important. The importance is that the MSE decreases.

Notice that the Mean Squared Error should not be used as the main guide for tuning. Tuning a system solely to achieve the lowest MSE possible may result in a system that is tuned very close to instability. Any change in the system, such as temperature or load, may cause the system to begin oscillating.

Adding the MSE to a Plot

The Mean Squared Error is included in default plots for control axes. To add the MSE to a custom plot, do the following:

In the Plot Template Editor, click New Quantity.
On the Advanced tab, in the Formula box, choose Mean Squared Error.
To choose the MSE for the position, velocity, pressure, or force, choose Standard Error Quantity and select the axis and quantity.
To choose the MSE for any other values, choose Custom Error Quantity and select the two registers.
Click Finish, then download the Plot Template to the RMC. The next plot captured for that plot template will include the MSE.

Mathematical Definition

The Mean Squared Error is the average of the square of the difference. The difference between each sample is squared, and the average of all the differences is the MSE.