Logo eprints

Simultaneous optimal control and discrete stochastic sensor selection

Bernardini, Daniele and Muñoz de la Peña, David and Bemporad, Alberto and Frazzoli, Emilio Simultaneous optimal control and discrete stochastic sensor selection. In: Hybrid Systems: Computation and Control. Lecture Notes in Computer Science, 5469 . Springer-Verlag Berlin Heidelberg, pp. 61-75. ISBN 978-3-642-00601-2 (2009)

Full text not available from this repository.


In this paper we present the problem of combining optimal control with efficient information gathering in an uncertain environment. We assume that the decision maker has the ability to choose among a discrete set of sources of information, where the outcome of each source is stochastic. Different sources and outcomes determine a reduction of uncertainty, expressed in terms of constraints on system variables and set-points, in different directions. This paper proposes an optimization-based decision making algorithm that simultaneously determines the best source to query and the optimal sequence of control moves, according to the minimization of the expected value of an index that weights both dynamic performance and the cost of querying. The problem is formulated using stochastic programming ideas with decision-dependent scenario trees, and a solution based on mixed-integer linear programming is presented. The results are demonstrated on a simple supply-chain management example with uncertain market demand.

Item Type: Book Section
Identification Number: 10.1007/978-3-642-00602-9_5
Projects: WIDE project
Funders: This work was partially supported by the European Commission under the HYCON Network of Excellence, contract number FP6-IST-511368, and under the WIDE project, contract number FP7-IST-224168
Subjects: Q Science > QA Mathematics
T Technology > TA Engineering (General). Civil engineering (General)
Research Area: Computer Science and Applications
Depositing User: Professor Alberto Bemporad
Date Deposited: 27 Jul 2011 08:32
Last Modified: 12 Oct 2011 14:23
URI: http://eprints.imtlucca.it/id/eprint/512

Actions (login required)

Edit Item Edit Item