Model Predictive Control for Production Machines: Learning and Stability Certificates

Add to your list(s) Download to your calendar using vCal

• Abhishek Dutta, University of Cambridge
• Friday 29 November 2013, 11:00-12:00
• 611.

If you have a question about this talk, please contact Eric C Kerrigan.

Traditional controllers in machines have important limitations. Firstly, in many cases it is impossible to optimally tune the controllers due to vaguely known dynamics of the machines. Furthermore, these controllers cannot cope with varying environmental conditions, which is inherent in practical situations. It will be demonstrated that, these drawbacks of traditional control algorithms, which result in suboptimal performances of the machines, can be resolved by incorporating learning behaviour into an advanced control strategy like MPC which solves a constrained optimal control problem, on-line, at each sampling instant, and the optimization is repeated at each subsequent time step, introducing feedback. However, industrial controllers are seldom designed ensuring stability, hence an attempt will be made towards certifying feasibility for the existing MPC controllers which would give guarantees to the industry about the controller stability under varying environmental conditions.

This talk is part of the Control and Power Seminars series.

This talk is included in these lists:

• 611
• Andrea Picciau's list
• COMMSP & CP list
• COMMSP Seminar
• Control and Power Seminars
• Featured talks
• SAP Talks

Note that ex-directory lists are not shown.