LinearMatrixInequalityOptions

Options & Configuration ยท AiDotNet.Models.Options

Configuration for the linear matrix inequality solver.

Inheritance: ModelOptions โ†’ LinearMatrixInequalityOptions

For Beginners

These settings control how long the solver searches, how large its first move is, and how carefully it estimates the most troublesome matrix direction. The defaults are safe for ordinary control problems.

How It Works

Reference: N. Z. Shor, Minimization Methods for Non-Differentiable Functions, Springer, 1985; S. Boyd et al., Linear Matrix Inequalities in System and Control Theory, SIAM, 1994.

Constructors

Constructor Summary
LinearMatrixInequalityOptions() Creates options with the documented defaults.
LinearMatrixInequalityOptions(LinearMatrixInequalityOptions other) Creates an independent copy of another LMI solver configuration.

Properties

Property Summary
double InitialStepSize { get; set; } Gets or sets the initial step length for the subgradient iteration.
double Margin { get; set; } Gets or sets how strictly inside the feasible set the search aims to land.
int MaxIterations { get; set; } Gets or sets the maximum number of subgradient iterations.
int PowerIterations { get; set; } Gets or sets the number of power iterations used to extract each eigenvector.