Co-design of embedded MPC controllers

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

• Andrea Suardi (Imperial College)
• Wednesday 20 February 2013, 11:00-12:00
• Level 9 Mahanakorn Lab., EE Dept..

If you have a question about this talk, please contact Grigorios Mingas.

In this talk, I propose a systematic and mathematically rigorous framework for the implementation of explicit model predictive control (MPC) algorithms on current and future embedded systems. The standard practice for controller design is that the control engineers design a controller assuming that its hardware implementation is a separate decoupled problem, although a high degree of coupling is indeed present. In fact, when tight performances (i.e. control, computing power, energy, etc.) and costs are required, the control algorithm has to be tailored for being implemented on a hardware platform. For instance, the choice the number of bits for data representation affects both the closed-loop performance (less bits lead to higher round-off errors and therefore less accurate control actions) and hardware specifications like power consumption, chip size, etc. I will discuss how this will be achievable by posing a multi-objective optimization problem, where the controller closed-loop and its algorithm implementation performances are minimized in a joint cost function. As example, I will show how the proposed co-design approach can be successfully applied to control a low power DC-DC converter.

This talk is part of the CAS Talks series.

This talk is included in these lists:

• Andrea Picciau's list
• CAS Talks
• Circuits and Systems Group: Internal Seminars
• Level 9 Mahanakorn Lab., EE Dept.

Note that ex-directory lists are not shown.