/// <summary>
/// Updates the parameter to be adapted acording to the given input and time/iteration.
/// Follows a discrete-time extremum seeking adaptation law.
/// </summary>
/// <param name="u">The input signal.</param>
/// <param name="t">The current time/iteration.</param>
/// <returns>The updated parameter value.</returns>
public float UpdateParameter(float u, float t)
{
// Clear states variable
states.Clear();
// Get gradient estimate
float[] duArr = estimator.Update(u, t);
// Update parameter estimate
float thetaBar = optimiser.Update(duArr);
// Add excitation dither
theta = thetaBar + dither.Update(t);
// Clamping
if (theta > thetaMax)
{
theta = thetaMax;
}
else if (theta < thetaMin)
{
theta = thetaMin;
}
// Add data to states
states.AddRange(duArr);
states.Add(thetaBar);
return(theta);
}
/// <summary>
/// Updates the parameter to be adapted acording to the given input and time/iteration.
/// Follows a discrete-time extremum seeking adaptation law.
/// </summary>
/// <param name="u">The input signal.</param>
/// <param name="t">The current time/iteration.</param>
/// <returns>The updated parameter value.</returns>
public float UpdateParameter(float u, float t)
{
// Clear states variable
states.Clear();
// Filter input signal
float uf = filter.Update(u);
// Get derivatives estimates
float[] duArr = estimator.Update(uf, t);
// Update parameter estimate
float thetaBar = optimiser.Update(duArr);
// Add excitation dither
theta = thetaBar + dither.Update(t);
// Clamping
if (theta > thetaMax)
{
theta = thetaMax;
}
else if (theta < thetaMin)
{
theta = thetaMin;
}
// Add data to states
states.Add(uf);
states.AddRange(duArr);
states.Add(thetaBar);
return(theta);
}