Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.

  

Semidefinite Programming for Stochastic Optimal Control

Author(s):

C. Stäheli
Conference/Journal:

Semester Thesis, HS14 (10485)
Abstract:

The objective of this semester project is to develop a general numerical tool to approximate the solution to generally nonlinear stochastic optimal control problem (SOC). by a semidefinite program (SDP), which is a convex optimization problem, and synthesize a controller for the stochastic system. The SOC problems are cast by relaxing the Bellan equation to an inequality and obtaining an equivalent infinite-dimensional linear program whose decision variable is the optimal value function on the SOC. Using polynomial basis functions and assuming polynomial problem data as well as knowledge of moments of the uncertainty and initial state distribution to approximate the value function, sum of squares programming technique yields a SDP. A greedy policy is optained from the suboptimal value function. he tool is validated on numerical examples to evaluate the permormance and limitations of the approach.

Supervisors: Maryam Kamgarpour, Tyler Summers, John Lygeros

Year:

2015
Type of Publication:

(13)Semester/Bachelor Thesis
Supervisor:

J. Lygeros

File Download:

Request a copy of this publication.
(Uses JavaScript)
% Autogenerated BibTeX entry
@PhdThesis { Xxx:2015:IFA_5097
}
Permanent link