S/sup 2/YSCODE: a software laboratory for stochastic systems control and decision algorithms, FORTRAN and MATLAB versions

Abstract

This paper presents an overview of the architecture and main features of S'YSCODE, a software laboratory for Stochastic SYStema Control and DEcision algorithms. There are two versions currently available, one in FOR-TRAN 77 and the other a fully transportable MATLAB version. Our principal aim have been to develop a userfriendly, interactive and transportable software laboratory. The current state of development makes S2YSCODE spe-The development environment for S'YSCODE has been the cluster of Sun SPARC stations in the Laboratory for Algorithmic Research, in the Systems k Industrial Engineering (SIE) Department at the University of Arizona. The organization of this report is as follows. In section 2 we summarize the SC/DS amenable to computational study. Section 3 summarizes the optimal control/de.cision algorithms. The architecture and commands implemented by SlYSCODE to date are discussed in section 4. cially well suited for pedagogical purposes. The current suited for large scale problems. 2 Stochastic Control/Decision focus is towarde the implementation of algorithma better Problems We consider discrete-time stochastic systems modeled aa a controlled Markov process [l), [2], [5], [IO], [Ill, [12]. The the dynamics are described by a system equation aa follows: 0-7803-1 968-0/94$4.00@1994 IEEE