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

Merge branch 'master' of https://git.frank-ebner.de/FHWS/IPIN2018

Browse Source
This commit is contained in:
toni
2018-10-21 00:37:15 +02:00
parent 9f77c69ad0 7c907c6bdd
commit ba81770054
6 changed files with 58 additions and 57 deletions
Download Patch File Download Diff File Expand all files Collapse all files

48
tex_review/chapters/experiments.tex
View File

@@ -8,54 +8,54 @@ The \del{\SI{2500}{\square\meter}} building consists of \SI{6}{} different level
Thus, the ceiling height is not constant over one floor and varies between \SI{2.6}{\meter} to \SI{3.6}{\meter}.
While most of the exterior and ground level walls are made of massive stones, the floors above are half-timbered constructions.
In the middle of the building is an outdoor area, which is only accessible from one side.
\add{The complete walkable indoor area for a visitor is \SI{2500}{\square\meter} in size.
\add{The total walkable indoor area for a visitor is \SI{2500}{\square\meter} in size.
Due to objects like exhibits, cabinets or signs not all positions within the building were freely accessible.}
For the sake of simplicity we did not incorporate such knowledge into the floorplan.
Thus, the floorplan consists only of walls, ceilings, doors, windows and stairs.
For the sake of simplicity we did not incorporate such knowledge into the floor plan.
Thus, the floor plan consists only of walls, ceilings, doors, windows and stairs.
It was created using our 3D map editor software (see fig. \ref{fig:mapeditor}) based on architectural drawings from the 1980s.
\add{The mesh is then created automatically, which only takes a few seconds to compute.}
\add{Our map editor is also used to automatically create the navigation mesh, which only takes a few seconds to compute.}
\begin{figure}[t]
\centering
\includegraphics[width=\textwidth]{gfx/apps/editor_light.png}
\caption{\add{The 3D map editor we developed to create the floorplans. This screenshot shows the ground level of the building. The window is split into toolbar (left), layers (upper right), parameters of current selection (lower right), drawing mode (upper center) and 3D view (lower center).}}
\caption{\add{The 3D map editor we developed to create the floor plans. This screenshot shows the ground level of the building. The window is split into toolbar (left), layers (upper right), parameters of current selection (lower right), drawing mode (upper center) and 3D view (lower center).}}
\label{fig:mapeditor}
\end{figure}
%wie haben wir die ap aufgehängt
\add{As described in section \ref{sec:wifi} we used \SI{42}{} WEMOS D1 mini to provide a \docWIFI{} infrastructure throughout the building.
\add{As described in section \ref{sec:wifi} we used \SI{42}{} WEMOS D1 mini to provide a \docWIFI{} signal coverage throughout the building.
The distribution of the beacons on ground floor can be seen in fig. \ref{fig:apfingerprint} (black dots) as well as the references (fingerprints) for optimization.
The position of the beacons were chosen depending on available power sources.
We have tried to have at least two beacons in one room and a third beacon visible in an approximate radius of 10 meters.
Due to the difficult architecture and the extremely thick walls of the museum, we decided on this procedure, which explains the rather unusual number of \SI{42}{} transmitters compared to modern buildings.
Another reason for the high number of beacons is that we did not want to analyze the quality of the Wi-Fi infrastructure for further improvements, as this can be a very time-consuming task.
Care was taken to have at least two beacons in each room and a third beacon visible in an approximate radius of \SI{10}{\meter}.
Due to the difficult architecture and the extremely thick walls of the museum, we decided on this procedure, which explains the rather large number of \SI{42}{} transmitters compared to modern buildings.
Another reason for the high number of beacons is that we did not want to analyze the quality of the Wi-Fi signal coverage for further improvements, as this can be a very time-consuming task.
In many areas of the building an improvement would not even be possible due to the lack of power sockets.
As an alternative, the beacons could also be operated using a battery, but we consider this approach less practicable, so we did not take this option.
The power sockets had different heights ranging from \SI{0.2}{\meter} to \SI{2.5}{\meter}.
So there were no prior requirements how to place a single beacon exactly and their position is thus similar to the sockets position.
To compensate that, battery powered beacons could be used but we consider this approach less practicable, so we did not take this option.
The power sockets are located at different heights ranging from \SI{0.2}{\meter} to \SI{2.5}{\meter}.
Consequently, there were no prior requirements on how a single beacon should be placed exactly and its position is dictated by the socket's position.
Considering all the above, the beacons were placed more or less freely and to the best of our knowledge.}
\add{A very similar approach was chosen for placing the fingerprints.
The positions of the fingerprints are set within our 3D map editor (see fig. \ref{fig:mapeditor}) software by simply dragging the fingerprinting icon on the desired position or by entering the position manually.
The reference points were placed every \SI{3}{\meter} to \SI{7}{\meter} from each other, however as can be seen in fig. \ref{fig:apfingerprint} not very accurately.
A perfect distance between the single points is not a crucial factor for the optimization and thus we consider such an accurate approach to be pointless.
\add{A similar approach was chosen for placing the fingerprints.
The positions of the fingerprints are set within our 3D map editor (see fig. \ref{fig:mapeditor}) software by dragging the fingerprinting icon on the desired position or by entering the position manually.
The reference points were placed every \SI{3}{\meter} to \SI{7}{\meter} from each other, however as can be seen in fig. \ref{fig:apfingerprint} not necessarily accurate.
As the optimization scheme does not require equally spaced reference points, doing so would result in superfluous effort.
Furthermore, it is not easy to adopt the exact position to take the reference measurements in the building later on.
Of course, this could be done with appropriate hardware (e.g. laser-scanner), but again this costs a lot of time, which in our opinion does not justify a presumably increased accuracy of some decimeters.}
Of course, this could be achieved with appropriate hardware (e.g. laser-scanner), but again, this requires more time and care, which in our opinion does not justify a presumably increased accuracy of some decimeters.}
\add{Summing up the above, the following initial steps are required to utilize our localization system to a building:
\begin{enumerate}
\item Acquiring a blueprint or architectural drawing of the building including at minimum the walls and stairs of the respective floors.
\item Based on this 2D drawing, the floorplan is created using our 3D map editor (cf. fig. \ref{fig:mapeditor}). This requires manual effort, comparable to software like Inkscape or FreeCAD.
\item If necessary, creating or improving the Wi-Fi infrastructure by plugging in beacons to available power sockets and write all MAC-addresses into a whitelist.
%\item Store floorplan and whitelist of MAC-addresses onto the smartphone.
\item Record the reference measurements based on the reference positions given in the floorplan.
\item Based on this 2D drawing, the floor plan is created manually using our 3D map editor (cf. fig. \ref{fig:mapeditor}), comparable to software like Inkscape or FreeCAD.
\item If necessary, create or improve the Wi-Fi infrastructure by plugging in beacons to available power sockets and compose a whitelist of MAC-addresses of the involved access points or beacons.
%\item Store floor plan and whitelist of MAC-addresses onto the smartphone.
\item Record the reference measurements based on the reference positions given in the floor plan.
\item The Wi-Fi model is optimized using the previously obtained reference measurements.
\item The navigation mesh is created automatically based on the before created floorplan as can be seen in fig. \ref{fig:museumMapMesh}.
\item The navigation mesh is created automatically based on the before created floor plan as can be seen in fig. \ref{fig:museumMapMesh}.
\end{enumerate}
For the building considered within this work, we were able to perform this steps in less then \SI{160}{\minute} by a person, which is familiar with the system, and the janitor of the museum.
Step 1 and 2 were conducted off-site.
The blueprint was initially provided by the director of the museum as digital photography.
Creating the floorplan including walls and stairs took us approximately \SI{40}{\minute} and is then stored onto the smartphone after creation.
Creating the floor plan including walls and stairs took us approximately \SI{40}{\minute} and is then stored onto the smartphone after creation.
Adding knowledge like semantic information such as room numbers would of course take additional time.
All other steps were performed on-site using our smartphone app for localization, which can be seen in fig. \ref{fig:yasmin}.
As the museum did not provide any Wi-Fi infrastructure, we installed the \SI{42}{} beacons as explained above.
@@ -82,7 +82,7 @@ In addition, the above steps do not require a high level of detail in their exec
\label{fig:simple}
\end{subfigure}
\caption{\add{The two mobile applications developed for Android. The localization app in (a) is used to record the Wi-Fi reference measurements based on the positions provided by the floorplan. In this screenshot the dialog for recording them is visible. The app also implements the here presented approach and can thus be used for localization. However, for the utilized experiments we used a simpler client (b) allowing for user input like a ground truth or activity button.}}
\caption{\add{The two mobile applications developed for Android. The localization app in (a) is used to record the Wi-Fi reference measurements based on the positions provided by the floor plan. In this screenshot the dialog for recording them is visible. The app also implements the here presented approach and can thus be used for localization. However, for the utilized experiments we used a simpler client (b) allowing for user input like a ground truth or activity button.}}
\label{fig:applications}
\end{figure}