Ausarbeitung/wiselib.tex

   1 \section{Wiselib}
   2 The \definition{Wiselib}\cite{wiselib} is a C++\index{C++} algorithm library for
   3 sensor networks, containing for example algorithms for routing, localization and
   4 time synchronization, and is strongly focused on portability and cross-platform
   5 development. In particular, it allows the user to develop applications that run
   6 on different hardware platforms without the need to change the code, and it
   7 strongly uses C++ templates to achieve that feature. Amongst the supported
   8 platforms are diverse sensor node platforms, like iSense, Contiki and TinyOS,
   9 but there are as well implementations for the diverse x86-compatible Personal
  10 Computer platforms, and the Shawn sensor network simulator.
  11
  12 \subsection{Architecture}
  13 \paragraph{Concepts and Models}
  14 Wiselib makes strong uses of \definition{concepts} and \definition{models} as
  15 central design objects. Concepts serve as an informal description of interfaces,
  16 only existent in documentation, defining expected parameters and types. Models
  17 however implement these interfaces in C++ code while fulfilling their
  18 specification. The Wiselib algorithms can in turn rely on the concepts as a
  19 generic specification, and take models as template parameters to use their
  20 functionality, so a function call will be immediately resolved to a specific
  21 model at compile time without the need for an additional function call as it is
  22 the case with virtual inheritance. Furthermore, this also allows usage on
  23 platforms which do not have support for C++, as the bytecode generated by the
  24 compiler does not include C++ specific extensions (no virtual function tables,
  25 and templates are resolved at compile time) and can be linked against any
  26 Standard C Library.
  27
  28 This makes cross-platform development easily possible. For example, to implement
  29 a routing algorithm, one can rely on the concept of a Radio to send and receive
  30 data packets, without needing to implement code specific to the used radio
  31 hardware. The users of that routing algorithm can now choose which radio model
  32 they want to use, according to their needs and the underlying hardware, provided
  33 that their radio model also implements the same Radio concept that the routing
  34 algorithm uses.
  35
  36 \begin{figure}
  37   \centering
  38   \includegraphics[width=.8\textwidth]{images/Wiselib-Arch.pdf}
  39   \caption{Wiselib architecture\label{fig:wiselib-arch}}
  40 \end{figure}
  41 Besides algorithms, and basic concepts and models, the Wiselib also consists of
  42 two other main parts: the internal interface and the external interface (see
  43 Figure \ref{fig:wiselib-arch}).
  44
  45 \paragraph{External Interface}
  46 The \definition{External Interface} provides access to the underlying \ac{OS},
  47 like iSense, Contiki, Shawn\ldots and defines concepts like a Radio or a Timer.
  48 Thus, the concepts are as generic as possible to match all supported operating
  49 systems and provide a light-weight abstraction to the underlying \ac{OS}. These
  50 concepts sometimes extend the generic concepts, for example there is a TxRadio
  51 which has the ability to set the transmission power on the radio. The models
  52 (for example an iSenseRadioModel or a ShawnTimerModel) implement these concepts
  53 on the specific operating system, and can be passed around as template
  54 parameters.
  55
  56 \paragraph{Internal Interface}
  57 The \definition{Internal Interface} defines concepts and models for data
  58 structures that can be used in algorithms. This allows to specialize for
  59 restricted platforms; for instance a wireless sensor node without dynamic memory
  60 management can use a static implementation of lists and other containers,
  61 whereas a full-grown desktop can use the dynamic implementation provided by the
  62 C++ \ac{STL}. For this purpose, the Wiselib also contains the
  63 \acused{pSTL}\definition{\acl{pSTL}} (\acs{pSTL}) which implements a
  64 subset of the \ac{STL} without the use of dynamic memory allocation.
  65
  66 \paragraph{Stackability}
  67 Another central design principle used in the Wiselib is
  68 \definition{stackability}, which describes the possibility to stack
  69 implementations of the same concept, thereby building a layered structure. For
  70 example, one could stack a cryptography algorithm on top of a radio model, which
  71 both implement the Radio concept. In this case, the cryptography layer doesn't
  72 have to know anything at all about the underlying implementation, as long as it
  73 can use the Radio concept of the underlying layer. And it can even provide the
  74 same interface to a possibly higher layer in order to provide transparent packet
  75 de- and encryption over the radio.
  76
  77 \subsection{Roomba Control}
  78 Even more interesting is the fact that the Wiselib includes code to control an
  79 iRobot Roomba\index{Roomba} over a serial interface, and getting access to its
  80 internal sensor data, using the \acl{ROI}\index{\acl{ROI}} mentioned earlier.
  81 For this purpose, it defines two concepts for Robot Motion:
  82
  83 \paragraph{TurnWalkMotion concept}\index{TurnWalkMotion (concept)}
  84 This concept represents a simple robot that can turn on the spot and walk
  85 straight, without automatic stopping.
  86 \begin{description}
  87   \item Types:
  88     \begin{description}
  89       \item[\code{velocity\_t}] Type for velocity measurement
  90       \item[\code{angular\_velocity\_t}] Type for angular velocity measurement
  91     \end{description}
  92   \item \todo{clearpage?} Methods:
  93     \begin{description}
  94       \item[\code{int turn(angular\_velocity\_t)}] turn the robot with a
  95         constant angular velocity
  96       \item[\code{int move (velocity\_t)}] move the robot straight with a
  97         constant velocity
  98       \item[\code{int stop()}] stop the robot
  99       \item[\code{int wait\_for\_stop()}] hold the execution until the robot has
 100         stopped
 101     \end{description}
 102 \end{description}
 103
 104 \paragraph{Odometer concept}\index{Odometer (concept)}
 105 This concept represents an Odometer which tracks motions over time.
 106 Whenever the object turns or moves, internal counters will adjust their
 107 guessing of the object's traveled distance and current orientation.
 108 \begin{description}
 109   \item Types:
 110     \begin{description}
 111       \item[\code{angle\_t}] Type for angle measurement
 112       \item[\code{distance\_t}] Type for distance measurement
 113     \end{description}
 114   \item Methods:
 115     \begin{description}
 116       \item[\code{angle\_t angle()}] return the current angle
 117       \item[\code{int reset\_angle()}] reset the angle of the object
 118       \item[\code{distance\_t distance()}] return the current distance
 119       \item[\code{int reset\_distance()}] reset the distance of the object
 120       \item[\code{int register\_state\_callback (T *obj)}] register a callback
 121         that gets called when the state changes
 122       \item[\code{int unregister\_state\_callback (int)}] unregister a
 123         previously registered callback
 124       \item[\code{int state()}] return the current state
 125     \end{description}
 126 \end{description}
 127
 128 \paragraph{ControlledMotion class}\index{ControlledMotion (class)}
 129 On top of the TurnWalkMotion and Odometer concepts builds the
 130 \definition{ControlledMotion} model. It takes implementations of each of these
 131 concepts as template parameters and extends the simple turn-and-walk paradigm by
 132 a temporal dimension, which let the robot stop after a specific time interval.
 133 In particular, it provides the following methods:
 134 \begin{description}
 135   \item[\code{int move\_distance(distance\_t, velocity\_t)}] move the robot
 136     straight by a given distance with a given velocity
 137   \item[\code{int turn\_about(angle\_t, angular\_velocity\_t)}] turn the robot
 138     about a given angle with a given angular distance
 139   \item[\code{int turn\_to(angle\_t, angular\_velocity\_t)}] turn the robot
 140     to a given orientation with a given angular distance
 141 \end{description}
 142
 143 The class first registers a callback function at the given Odometer instance,
 144 and then uses its distance and angle values to control the robot over the
 145 TurnWalkMotion instance. Everytime the state of the robot changes (i.~e. new
 146 data from its sensors are received), it compares the new actual values with the
 147 target values given by the user through the functions above, and if the actual
 148 values exceed the target values, the robot is stopped.
 149
 150 \paragraph{Underlying Roomba Implementation}
 151 The actual communication with the Roomba is done in the \definition{RoombaModel}
 152 class. It implements the aforementioned TurnWalkMotion \index{TurnWalkMotion
 153 (concept)} and Odometer \index{Odometer (concept)} concepts and therewith allows
 154 the interaction with a ControlledMotion instance. In particular, it manages the
 155 serial communication with the Roomba and translates the function calls
 156 \code{turn()}, \code{move()} and \code{stop()} of the TurnWalkMotion concept to
 157 the according parameters for the \ac{ROI} \cmd{Drive} command, reads a subset of
 158 the Roomba's sensors and presents the sensor data to the user, and, while
 159 implementing the Odometer concept, calculates the covered distance and angle
 160 from the Roomba's right and left wheel rotations.
 161
 162 The sensor data is read from the Roomba using the \cmd{Stream} command on the
 163 \ac{ROI}, which results in a sensor data packet (see Section
 164 \ref{sec:roi-stream-packet}) every 15~ms, and $66.67$ packets per second. At a
 165 speed of 19,200 baud in mode \ac{8N1}, the maximum size of a data packet is
 166 $19,200 \div (66.67 \times 9) = 32$ byte, so at the moment only the sensor
 167 packets \emph{encoder counts left/right} (IDs~\magicvalue{0x2b},
 168 \magicvalue{0x2c}, 2+2 bytes), \emph{battery voltage/current/charge/capacity}
 169 (IDs~\magicvalue{0x16}, \magicvalue{0x17}, \magicvalue{0x19},
 170 \magicvalue{0x1a}, 2+2+2+2 bytes) are streamed, which add up to 18 data bytes +
 171 3 header/checksum bytes. There has currently been no success yet in
 172 communicating at the higher speed of 115,200 baud.
 173
 174 Also there has been research to use the distance and angle values that the
 175 Roomba itself provides (sensor packet~IDs \magicvalue{0x13} and
 176 \magicvalue{0x14}). However, according to the \ac{ROI} Specification these
 177 values are integer values, and the value is reset to zero every time it is read.
 178 By using the \cmd{Stream} command on the \ac{ROI} and therefore reading these
 179 values every 15~ms, the Roomba cannot move fast enough to increment these values
 180 to \magicvalue{1}, so everytime \magicvalue{0} is read. It is obvious that
 181 these sensor values are not suited for such rapid evaluations, and can only be
 182 used for larger distances and angles. Nevertheless, since the Wiselib Roomba
 183 control needs to keep track of the Roomba's current position and orientation as
 184 fast as possible to maintain a certain accuracy in movement, the distance and
 185 angle values as provided by the Roomba itself cannot be used. On the other hand,
 186 working with the Roomba's wheel encoder counts (sensor packet~IDs
 187 \magicvalue{0x2b} and \magicvalue{0x2c}) has proven itself quite acceptable.
 188 After a few test runs, the number of encoder counts per~mm for straight
 189 walks turned out as $2.27$, and for turning on the spot, $2.27 \times 115 =
 190 261.05$ encoder counts per radian, which is the number of encoder counts per~mm
 191 multiplicated with half the Roomba's wheelbase. So when new sensor data is read
 192 each 15~ms, the RoombaModel implementation calculates the Odometer
 193 \index{Odometer (concept)} distance and angle from these values.
 194
 195 \todo{cite Wisebed book chapter on Roomba code}