Implementation aspects of embedded MPC with fast gradient method

In this paper we investigate the use of the Model Predictive Control (MPC) technique on a low power embedded computing platform. The control approach uses a quadratic optimization problem to compute the optimal control signal. The problem is solved subject to a linear model of the system and the phy...

Full description

Bibliographic Details
Main Authors: Novák, Jakub, Chalupa, Petr
Format: Article in Journal/Newspaper
Language:English
Published: North Atlantic University Union (NAUN) 2014
Subjects:
Embedded systems
Fast gradient method
Model predictive control
Multiple models
North Atlantic
Online Access:http://publikace.k.utb.cz/handle/10563/1005654
Description
Summary:In this paper we investigate the use of the Model Predictive Control (MPC) technique on a low power embedded computing platform. The control approach uses a quadratic optimization problem to compute the optimal control signal. The problem is solved subject to a linear model of the system and the physical limitations of the system. The optimization problem is solved online using the Fast Gradient method. The proposed controller has been implemented on a Stellaris Launchpad board with ARM Cortex processor. By means of two simulation studies we detail the software and the hardware aspects concerning a fast realtime MPC implementation. In the first example linear MPC is used for stabilization of a quadrotor model. In the second example nonlinear pH neutralization plant is controlled using fuzzy MPC algorithm. © 2014 North Atlantic University Union. All rights reserved.

Similar Items