Ausarbeitung/implementation.tex

   1 \chapter{Implementation}
   2
   3 \todo{?}
   4 This chapter describes the implementation that was used for the aforementioned
   5 experiments. It consists of the measuring program itself, as well as several
   6 scripts to help with analysis of the measured data.
   7
   8 \section{Measuring}
   9 \label{sec:impl:measuring}
  10
  11 C++ code in wiselib/trunk/pc\_apps/roomba\_tests
  12
  13 three single applications with same base: roomba\_test (main.cc),
  14 mean\_correction\_test (mean\_correction.cc), soft\_start\_test
  15 (soft\_start.cc).
  16
  17 user interface: Qt.
  18
  19 manual tests (user is asked for new values every time) or automated tests (user
  20 only asked for measured values).
  21
  22 1) open UART connection
  23
  24 2) register state callback for getting roomba sensor data
  25
  26 3) while(input values or cancel) { drive() / turn() }
  27
  28 4) drive() / turn() use wiselib::ControlledMotion<OsModel,
  29 wiselib::RoombaModel> for moving specified angle/distance; and ask for measured
  30 values and write to external log file, including battery status, roomba/wiselib
  31 internal angles/distances (ticks), svn revision, floor type, roomba ID
  32
  33 additionally, mean correction test uses
  34 CorrectedMeanMotion from corrected\_mean\_motion.h, implements same concept like
  35 wiselib::ControlledMotion and takes care of target value by using the
  36 calculated fit function .
  37
  38 additionally, soft start/stop test uses
  39 SoftStartMotion from soft\_start\_motion.h, implements same
  40 concept like wiselib::ControlledMotion and takes care of increasing/decreasing
  41 velocity via timer.
  42
  43 \section{Evaluation}
  44 \label{sec:impl:eval}
  45 \todo{}
  46
  47 bash/perl scripts in wiselib/trunk/pc\_apps/roomba\_tests/logs, using gnuplot
  48
  49 graph.sh: create 3d plots (input value, input velocity, measured value) from
  50 original behvaiour data, including fit function calculated by GNU R statistics
  51 software, for {carpet floor, laminate floor} $\times$ {drive straight, turn on
  52 spot}
  53
  54 graph-mean.sh: do the same for mean correction data, including fit function from
  55 original data
  56
  57 graph-soft.sh: do the same for soft start/stop data, including fit function from
  58 original data
  59
  60 graph-mean-soft.sh: 3d plot with mean correction and soft start/stop data, for
  61 comparison of both
  62
  63 graph-errorlines.sh: create 2d plots input value -> measured value, with
  64 multiple velocities in each graph. also split graphs up for {carpet floor,
  65 laminate floor} $\times$ {drive straight, turn on spot}. no fit function.