this is for holding javascript data
Volker Strobel edited chapter_Introduction_label_chap_introduction__.tex
almost 8 years ago
Commit id: 4812a87da81f1d13ec1485801c5a8c70a62aece1
deletions | additions
diff --git a/chapter_Introduction_label_chap_introduction__.tex b/chapter_Introduction_label_chap_introduction__.tex
index a928cf8..14a7150 100644
--- a/chapter_Introduction_label_chap_introduction__.tex
+++ b/chapter_Introduction_label_chap_introduction__.tex
...
filter (ii) software for augmenting an image with synthetic views,
(iii) a script for evaluating a map based on histograms and
corresponding $x,y$-positions.
\section{Research Questions}
\label{sec:researchquestions}
The research questions addressed in this thesis are:
\begin{itemize}
\item Can 2D positions be estimated in real-time using a machine
learning approach on a limited processor in a modifiable indoor
environment?
%\item Is accurate real-world localization regression possible when the
% training data comprises synthetic data only?
\item Can we predict the suitability of a given map for the proposed
localization approach?
\end{itemize}
\section{Outline}
\label{sec:outline}
The remainder of this thesis is structured as follows.
Chapter~\ref{chap:relatedwork} surveys existing indoor localization
approaches related to this thesis. In Chapter~\ref{chap:methods}, the
developed texton-based approach is presented and its components, the
$k$NN algorithm and the particle filter, are introduced. Details about
the synthetic data generation tool and map evaluation technique are
also given. Chapter~\ref{chap:analysis} describes the setup of the
on-ground and in-flight experiments. The results of the experiments
are presented in Chapter~\ref{chap:results} and discussed in
Chapter~\ref{chap:discussion}. Finally, we draw our conclusions and
indicate future research directions in Chapter~\ref{chap:conclusion}.