Model Predictive Control Toolbox

Customizing Constraints and Weights

Model Predictive Control Toolbox provides several tools to help you optimize controller performance by customizing controller constraints and weights.

Adjusting Weights with the Tuning Advisor

The toolbox provides the Tuning Advisor, which guides you in setting weights to improve controller performance. You can use the Tuning Advisor to:

  • Select a cost function that measures the difference between a reference signal and measured plant output, and compute the cost function value for the baseline design
  • Compute sensitivities of the cost function to individual weights
  • Determine whether individual weights should be increased or decreased to improve controller performance
  • Adjust the weights and recompute the cost function value

By repeating this interactive process, you can systematically adjust controller weights to optimize controller performance.

Weight Tuning for Model Predictive Controllers 7:24
Use Tuning Adviser to adjust model predictive controller weights to improve controller performance.

Analyzing Constraints and Weights for Potential Run-Time Failures

The product provides a diagnostic function to detect potential stability and robustness issues with your model predictive controller, such as:

  • The model predictive controller or the closed-loop system is unstable.
  • The quadratic programming (QP) optimization problem is ill-defined with an invalid Hessian matrix.
  • Zero steady-state offset cannot be achieved.
  • Hard and soft constraint settings may lead to infeasible optimization problems at run-time.

You can use this diagnostic tool to adjust controller weights and constraints during controller design to avoid run-time failures.

Results of diagnostic tests for potential model predictive controller run-time failures.
Results of diagnostic tests for potential model predictive controller run-time failures.
Next: Control Plants Over a Range of Operating Conditions

Try Model Predictive Control Toolbox

Get trial software

Diseño de sistemas de control multivariable para robótica

View webinar