RVNN

Name:

Control method with neural networks
Neuro-Control

Event type:

Lecture

Content:

Today, neural networks are increasingly used as controllers in practice. A successful application requires the knowledge of the basics of such methods. The aim of the course is to teach the theoretical basics for the use of neural networks as controllers in a controlled system. After successful completion of the module, students should be able to perform and evaluate the use of neural networks as controllers in an exemplary manner.

From the content:

• Control structures.

• Limits of conventional control with linear controllers.
• Requirements in practice: nonlinearity, self-adjustment, ongoing adaptation.
• Neural networks as models and as controllers:
  architectures and learning procedures:
  • System identification;
  • direct inverse control;
  • Control with internal model;
  • Feedback linearization;
  • Closed-loop control with feedforward;
  • Optimal control.
• off-line and on-line use.
• Stability.

Target audience:

Students in the 1st and 2nd level of study from the 5th semester onwards.

Scope:

2 SWS lecture, 2 SWS exercise, 6 CP

Dates:

The lecture takes place in the winter semester.

Documents:

Announcement in the lecture

Certificate of achievement:

Written exam

Lecturer:

• Dr.-Ing. Mohamed Ayeb