\relax 
\newlabel{beginFedorova}{{5}{31}}
\citation{blanke:1999}
\citation{deb:2001}
\citation{maybeck:1982}
\@writefile{toc}{\contentsline {paper}{\hbox to 2.1em {\textbf  {3}} \textit  {Maria Fedorova}.\tmspace  +\thickmuskip {.2777em} STOCHASTIC CONTROL ITERATED DESIGN OPTIMIZATION USING GENETIC ALGORITHMS}{32}}
\@writefile{toc}{\contentsline {section}{\numberline {1}Introduction}{32}}
\newlabel{intro}{{1}{32}}
\@writefile{toc}{\contentsline {section}{\numberline {2}Problem statement}{32}}
\newlabel{problem}{{2}{32}}
\newlabel{x}{{1}{33}}
\newlabel{y}{{2}{33}}
\newlabel{hatx-}{{3}{33}}
\newlabel{hatx+}{{4}{33}}
\newlabel{nu}{{5}{33}}
\newlabel{u}{{6}{33}}
\newlabel{effector}{{7}{33}}
\citation{basseville:1993}
\citation{semoushin:2004-2}
\citation{semoushin:2003}
\citation{grewal:2001}
\citation{maybeck:1982}
\@writefile{lof}{\contentsline {figure}{\numberline {1}{\ignorespaces The general framework for change point detection and filter selection. {\sl  Legend:} $\@EuScript {SP}$ stands for system population, each member of $\@EuScript {SP}$ being in reality the system (\ref  {x})--(\ref  {y}) with a particular value $\theta _{k}\in \Theta $, $k\in \{0,1,\dots  , K\}$, $K\in {\@mathbb {N}}$. \tmspace  +\thinmuskip {.1667em} $\@EuScript {FP}$ stands for filter population of size $K$; each member $\@EuScript {F}_{k}$ of $\@EuScript {FP}$ is the Kalman filter $\@mathrm {KF}_{\theta _{k}}$ being the optimal one with respect to system (\ref  {x})--(\ref  {y}) with a particular value $\theta _{k}\in \Theta $. \tmspace  +\thinmuskip {.1667em} For each single $\@mathrm {KF}_{\theta _{k}}$, decision $d_{k}=0$ is made in support of $\@mathrm {KF}_{\theta _{k}}$ or decision $d_{k}=1$ is made in refusal of $\@mathrm {KF}_{\theta _{k}}$. $\@EuScript {FS}$ stands for filter selection based on the all partial decisions $\{d_{k}\}$. \tmspace  +\thinmuskip {.1667em} The stopping rule generated by the $\@EuScript {DG}$ is defined by the alarm time $t_{\@mathrm {a}}=\qopname  \relax m{min}\mathopen {\setbox \z@ \hbox {\frozen@everymath \@emptytoks \mathsurround \z@ $\nulldelimiterspace \z@ \left \{\vcenter to1.5\big@size {}\right .$}\box \z@ } t_{k}: \tmspace  +\thickmuskip {.2777em} \mathopen {\setbox \z@ \hbox {\frozen@everymath \@emptytoks \mathsurround \z@ $\nulldelimiterspace \z@ \left (\vcenter to\@ne \big@size {}\right .$}\box \z@ } \exists \kappa : \ d_{\kappa }(t_{k})=0 \tmspace  +\thickmuskip {.2777em}\tmspace  +\medmuskip {.2222em} \& \mathop {\forall }\limits _{j\not =\kappa } d_{j}(t_{k})=1 \mathclose {\setbox \z@ \hbox {\frozen@everymath \@emptytoks \mathsurround \z@ $\nulldelimiterspace \z@ \left )\vcenter to\@ne \big@size {}\right .$}\box \z@ } \mathclose {\setbox \z@ \hbox {\frozen@everymath \@emptytoks \mathsurround \z@ $\nulldelimiterspace \z@ \left \}\vcenter to1.5\big@size {}\right .$}\box \z@ }$.\tmspace  +\thinmuskip {.1667em} $\@EuScript {FE}$ stands for feedback effector to implement the desired action (\ref  {effector}) in the system feedback $\@EuScript {SF}$. The only filter selected and uploaded to $\@EuScript {SF}$ is $\@EuScript {F}_{\kappa }$. }}{34}}
\newlabel{fig2}{{1}{34}}
\citation{semoushin:1985}
\citation{semoushin:1994}
\citation{semoushin:2000}
\citation{caines:1988}
\@writefile{toc}{\contentsline {section}{\numberline {3}Decision generator for detection}{35}}
\newlabel{detection}{{3}{35}}
\newlabel{API}{{8}{35}}
\newlabel{zeta}{{9}{35}}
\newlabel{d}{{10}{35}}
\newlabel{hatzeta}{{11}{35}}
\@writefile{toc}{\contentsline {section}{\numberline {4}GA based filter selection}{35}}
\newlabel{genetic}{{4}{35}}
\citation{maybeck:1982}
\citation{kristinsson:1992}
\@writefile{toc}{\contentsline {section}{\numberline {5}Experimental results}{39}}
\newlabel{experiment}{{5}{39}}
\@writefile{lof}{\contentsline {figure}{\numberline {2}{\ignorespaces Genetic algorithm performance: $\mathaccentV {hat}05EK_{\@mathrm {p}}$ and $\mathaccentV {hat}05EK_{\@mathrm {r}}$.}}{40}}
\newlabel{GA-1-PRM}{{2}{40}}
\@writefile{lof}{\contentsline {figure}{\numberline {3}{\ignorespaces Genetic algorithm performance: $\mathaccentV {hat}05EK_{\@mathrm {p}}$.}}{40}}
\newlabel{GA-1-P}{{3}{40}}
\@writefile{lof}{\contentsline {figure}{\numberline {4}{\ignorespaces Integral Percent Error for the controlled plant. Comparison of three methods: SSA, SLS, GA.}}{41}}
\newlabel{IPE-P}{{4}{41}}
\@writefile{lof}{\contentsline {figure}{\numberline {5}{\ignorespaces Integral Percent Error for the reference signal model. Comparison of SSA, SLS and GA.}}{41}}
\newlabel{IPE-RM}{{5}{41}}
\bibcite{basseville:1993}{1}
\bibcite{blanke:1999}{2}
\bibcite{caines:1988}{3}
\bibcite{deb:2001}{4}
\bibcite{grewal:2001}{5}
\bibcite{kristinsson:1992}{6}
\bibcite{maybeck:1982}{7}
\bibcite{semoushin:2003}{8}
\bibcite{semoushin:1985}{9}
\@writefile{toc}{\contentsline {section}{\numberline {6}Conclusions}{42}}
\newlabel{conclusions}{{6}{42}}
\bibcite{semoushin:2000}{10}
\bibcite{semoushin:2004-2}{11}
\newlabel{endFedorova}{{11}{43}}
\@setckpt{fedorova}{
\setcounter{page}{44}
\setcounter{equation}{11}
\setcounter{enumi}{5}
\setcounter{enumii}{6}
\setcounter{enumiii}{0}
\setcounter{enumiv}{11}
\setcounter{footnote}{1}
\setcounter{mpfootnote}{0}
\setcounter{part}{0}
\setcounter{section}{6}
\setcounter{subsection}{0}
\setcounter{subsubsection}{0}
\setcounter{paragraph}{0}
\setcounter{subparagraph}{0}
\setcounter{figure}{5}
\setcounter{table}{0}
\setcounter{papernumber}{3}
\setcounter{chapter}{0}
\setcounter{parentequation}{0}
\setcounter{subfigure}{0}
\setcounter{lofdepth}{1}
\setcounter{subtable}{0}
\setcounter{lotdepth}{1}
\setcounter{float@type}{4}
\setcounter{LT@tables}{0}
\setcounter{LT@chunks}{0}
\setcounter{remarka}{0}
\setcounter{theo}{0}
\setcounter{coro}{0}
\setcounter{alg}{0}
\setcounter{defn}{0}
\setcounter{lem}{0}
\setcounter{rusremarka}{1}
\setcounter{rustheo}{1}
\setcounter{ruscoro}{0}
\setcounter{rusalg}{1}
\setcounter{rusdefn}{0}
\setcounter{ruslem}{0}
}