FHWS/IPIN2018
Archived
4
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

first draft introduction

Browse Source
This commit is contained in:
toni
2018-03-22 14:38:41 +01:00
parent 74a2f7264e
commit 6f13ac9165
3 changed files with 43 additions and 52 deletions
Download Patch File Download Diff File Expand all files Collapse all files

77
tex/chapters/introduction.tex
View File

@@ -1,68 +1,43 @@
\section{Introduction}
Setting up a localization solution for a building is a challenging and time-consuming task, especially in environments that are not build with localization in mind or do not provide any wireless infrastructure or both.
Such scenarios are of special interest when old or even historical buildings serve a new purpose such as museums, shopping malls or retirement homes.
Setting up a localization solution for a building is a challenging and time-consuming task, especially in environments that are not build with localization in mind or do not provide any wireless infrastructure or even both.
Such scenarios are of special interest when old or historical buildings serve a new purpose such as museums, shopping malls or retirement homes.
In terms of European architecture, the problems emanating from these buildings worsen with age.
Such buildings are often full of nooks and crannies, what makes it hard for dynamical models using any kind of pedestrian dead reckoning (PDR). Here, the error accumulates not only over time, but also with the number of turns and steps made \cite{}.
There is also a higher chance of detecting false or misplaced turns, what can cause the position estimation to lose track or get stuck within a demarcated area.
Thus, this paper presents a very robust but realistic movement model using a navigation mesh based on triangles.
\todo{into all three dimensions \\what allows for very small map sizes}
In the scope of this work, we deployed an indoor localization system to a 13th century building.
The first 300 years the building was used as a convent, after that it had different functions ranging from a granary to an office for Bavarian officials.
Over this period, the building had major construction measures and was extended several times.
Since 1936, the \SI{2500}{m$^2$} building acts as a museum of the medieval town Rothenburg ob der Tauber \cite{Rothenburg}.
In localization systems using a sample based representation, like particle filters, the above mentioned problems can further lead to more advanced problems like sample impoverishment \cite{}.
Such buildings are often full of nooks and crannies, what makes it hard for dynamical models using any kind of pedestrian dead reckoning (PDR). Here, the error accumulates not only over time, but also with the number of turns and steps made \cite{Ebner-15}.
There is also a higher chance of detecting false or misplaced turns, what can cause the position estimation to lose track or get stuck within a demarcated area.
Thus, this paper presents a very robust but realistic movement model using a three-dimensional navigation mesh based on triangles.
%In addition, this allows for very small map sizes, consuming little storage space.
In localization systems using a sample based representation, like particle filters, the above mentioned problems can further lead to more advanced problems like sample impoverishment \cite{Fetzer-17} or multimodalities \cite{Fetzer-16}.
Sample impoverishment refers to a situation, in which the filter is unable to sample enough particles into proper regions of the building, caused by a high concentration of misplaced particles.
Within this work we present a simple yet efficient method that enables a particle filter to fully recover from sample impoverishment.
\todo{exact estimation fehlt hier noch}
We also use a novel approach for finding an exact estimation of the pedestrian's current position by using a rapid computation scheme of the kernel density estimation.
Many historical buildings, especially bigger ones like castles, monasteries or churches, are built of massive stone walls and have annexes from different historical periods out of different construction materials.
This makes it hard for methods using received signal strengths (RSS) from Wi-Fi or Bluetooth, due to a high signal attenuation between different rooms.
Many unknown quantities like the walls definitive material or thickness, makes it expensive to determine important parameters as the signals depletion over distance.
Additionally, most wireless approaches adapt a line-of-sight assumption, the performance will be even more limited due to the irregularly shaped spatial structure of such buildings.
This leads to problems for methods using received signal strengths (RSS) from Wi-Fi or Bluetooth, due to a high signal attenuation between different rooms.
Many unknown quantities like the walls definitive material or thickness make it expensive to determine important parameters, e.g. the signals depletion over distance.
Additionally, most wireless approaches adapt a line-of-sight assumption.
Thus, the performance will be even more limited due to the irregularly shaped spatial structure of such buildings.
Our approach tries to avoid those problems.
We distributing a small number of simple and cheap Wi-Fi beacons over the whole building and instead of measuring their position inside the building, we use a optimization scheme based on some reference measurements.
We distribute a small number of simple and cheap Wi-Fi beacons over the whole building and instead of measuring their position, we use a optimization scheme based on some reference measurements.
A optimization scheme also helps against inaccuracies like wrong positioned access points or fingerprints caused by outdated or inaccurate building plans.
It is obvious, that this could be solved by re-measuring the building, however this is a very time-consuming process requiring specialist hardware and in most cases a surveying engineer.
It is obvious, that this could be solved by re-measuring the building, however this is a very time-consuming process requiring specialist hardware and a surveying engineer.
However, this is contrary to most costumers expectations of a fast to deploy and low-cost solution.
In addition, this is not only a question of costs incurred, but also for buildings under monumental protection, what does not allow for larger construction measures.
Thus a highly flexible, robust and scalable system is needed to deal with such conditions, especially with the claim of a universal solution in multiple different environments.
Smartphone-based solution was choosen...
To tackle the challenges above, a novel indoor localization approach is presented within this paper and then deployed to a 13th century building.
The first 300 years the building was constructed and used as a convent, after that it had different functions ranging from a granary to an office for Bavarian officials.
Since 1936, the \SI{2500}{m$^2$} building acts as a museum of the medieval town Rothenburg ob der Tauber \cite{}.
To sum up, this work presents a smartphone-based localization system using a particle filter to incorporate different probabilistic models.
We omit time-consuming approaches like classic fingerprinting or measuring the exact positions of access-points by using a simple optimization scheme.
The pedestrian's movement is modeled realistically on a navigation mesh.
A barometer based activity recognition enables to go into the third dimension and problems occurring from multimodalities and impoverishment are taken into account.
We believe that by utilizing our localization approach to such a challenging scenario, it is possible to prove its flexibility, robustness and manageable effort.
The goal of this work is to propose a fast to deploy and low-cost localization solution, that provides reasonable results in a high variety of situations.
Consequently, we believe that by utilizing our localization approach to such a challenging scenario, it is possible to prove those characteristics.
It should finally be mentioned, that the here presented work is an updated and highly re-factored version of the winner of the smartphone-based competition at IPIN 2016 \cite{Ebner-15}.
Finally, the here presented work is an updated and highly refactored version of the winner of the smartphone-based competition at IPIN 2016 \cite{}.
\cite{Ebner-17}
wifi line of sight modelle versagen bei dicken wänden
\todo{stark aufs wlan eingehen in den experimenten, das wir hier neue modelle brauchen, weil die line of sight annahme einfach zu schwach ist bei solchen wänden.}
max. 1 Seite
\begin{itemize}
\item Deploying and indoor localisation system in the wild is not an easy task. especially in environments not planned for "zurechtfinden".... within this work we investigate the capabilities of a localization approach within a 14th century kloster, now blabal as a museum. the 2500 m2 building has no digital infrastructure like ethernet, wifi or bluetooth. within the last 600 years the building durchging einige mayor baumaßnahmen, wie das hinzufügen von stockwerken oder ganzen gebäudetrackte. leaving the build to be ein bunter mix (patchwork) unterschiedliche epochen, architekturen und verwendeter baumateriallien. (mehr dazu dann im indoormap kapitel)
\item im museums umfeld bringt es dies und das weil..
\item knappe kassen und kaum infrastruktur in den gebäuden
\item viele museun sind in sehr alten gebäuden unter gebracht mit historisch gewachsener infrastruktur, welche nicht für lokalisierung gemacht wurde
\item es braucht eine kostengünstige lösung, welche dem anspruch eines museums gerecht wird.
\item die wartung muss sehr gering ausfallen, da personal teuer
\item keine spezialhardware, die besucher sollen mit eigenen geräten erkunden können. also smartphone-based.
\item diese arbeit stellt daher ein smarthone-based lokalisierungssystem vor.
\item zur lokalisierung wird neben pdr eine simple infrastruktur aus wifi beacons genutzt, deren position über einige wenige fingerprints geschätzt werden.
\end{itemize}