Tracker Server Users Guide

Application Overview

1. It allows the user to graphically construct the physical environment that the positioning system will service. This includes the dimensions of the area that the positioning system will service as well as the location and headings of each of the cameras.

2. It uses the physical light information from the ITMC Tracker application and combines it with the physical attributes of the environment and computes the position of an infrared light source within the environment.

3. It serves all of this positioning information in two ways, visually and over the network. Visually, the position of the light source will be shown within the graphical environment constructed earlier. Client applications will be able to connect TrackerServer over the network and request positioning information.

Installing and running the application

1. Create the application directory.( EX C:\TrackerServer )
2. Download TS.ZIP to the directory created in step 1.
3. Unzip TS.ZIP.
4. Move or copy all of the DLL's to the windows system32 directory.
5. Create a shortcut to the TrackerServer.EXE file located in the directory created in step 1.
6. Start TrackerServer by double clicking on the shortcut.

Creating the environment

• Environment Description

  The TrackerServer environment consists of two objects rooms and cameras. The room object represents the room, or rooms, in which the indoor positioning system will be installed and used. The camera object represents the infrared cameras that are utilized by the ITMC Tracker system. The purpose of the TrackerServer environment is to emulate as closely as possible the exact dimensions and features of the room, or rooms, in which the IMTC Tracker system in installed in. The coordinate system that is used within TrackerServer is arbitrary. For example, When picking the origin for the first room you can choose a coordinate system where the one corner of the room is 0,0 or you can choose a coordinate system where one corner of the building containing the room is 0,0. It doesn't matter, just choose one in the beginning and stick with it. X increases as you move to the right. Y increases as you move up. As for the positions of the cameras, the Location of the camera is in the "world" coordinate system you choose, not within the coordinate system of a room (unless your room contains the origin of the "world" coordinate system).

  

• Adding and Editing a room

  To add a room, select the Add Room button from the toolbar. This button has a small cube on it.

  

  Specify the name of the Room, the origin of the room in "world" coordinates, and the dimensions of the room in feet.

  WARNING: Make sure the camera is located inside of a room.

  Select OK. The room should appear on the screen.

  

  To edit a room double click anywhere inside of the room. The room properties dialog should appear displaying the properties of the room. Make the necessary changes and select OK.

• Adding and Editing a camera

  To add a camera, select the Add Camera button from the toolbar. This button has a small camera on it.

  

  Specify the location in "world" coordinates. The heading of the camera is specified in degrees with respect to the X,Y, and Z axis. The X axis is to the right, the Y axis straight up, and the Z axis comes out of the screen toward you. Specify a rotation for the X,Y, and Z axis in degrees.

  NOTE: Only the Z Axis is being used in computations.

  Specify the camera number. This is the IMTC Camera number. Specify the physical field of view for the camera in degrees. Specify the Tracker Camera Extents. These extents are encoded within the Tracker application. Currently the Top Left of each camera used by the Tracker application is 0,0 and the Bottom Right is 245,235. Select OK. The camera should be displayed in the correct position, with the correct orientation and field of view. Also the camera number should superimposed on the camera.

  

  To edit a cameradouble click anywhere near the camera. The camera properties dialog should appear displaying the properties of the camera. Make the necessary changes and select OK.

• Saving the environment

  To save the environment, simply select Save from the File pulldown menu or click on the Save button on the toolbar. Specify the directory and filename and select OK.

• Opening a previously saved environment

  To open a previously saved environment select Open from the File pulldown menu or select the Open button from the toolbar. Locate the file and either double click on it or select it and hit OK. The environment should be loaded and displayed on the screen.

Using the Server Component

• Server component description

  The server component of the application connects to the IMTC Tracker application using TCP/IP sockets and retrieves all of the camera information. Then using this information it computes the position of the object and renders it on the screen. If the object is only visible to one camera, the line of sight is rendered as a blue line, and the position computed is the intersection of the line of sight with an intersecting wall of the room. If the object is visible to two cameras, then the object is rendered as a red circle, and the position computed is the intersection of the two camera lines of sight.

  NOTE: The application currently tracks only one object.

  The position of the object is also made available to client applications over a socket. Client applications can connect to the server and request the position of an object using the TrackerServer positioning protocol. See Implementing a client.

• Demo Mode

  The server component can be used in a demo mode without having to connect to the ITMC Tracker application. This is useful for debugging and demo purposes. To toggle the Demo mode on and off, select Demo mode from the Options pulldown menu. In Demo mode the, the application simulates input from camera 0 and camera 4 in the form of cosine waves. The values produced bounce back and forth between the Tracker Camera Extents.

• Starting the Server Component

  To start the server component, click on the Play button on the toolbar. This button has a small triangle on it. The position of the object will be periodically rendered to the screen in the form of a blue line or a red circle. A blue line means the object is only visible by one object and this line represents the line of sight with respect to the camera. A red circle means the object is visible by at least two cameras and this circle is the intersection of their individual lines of sight. At this point client applications can connect to TrackerServer and retrieve the position of the object.

  

  

• Stopping the Server Component

  To stop the server component, click on the Stop button on the toolbar. This button has a small square on it. At this point the application will drop the socket connection to the IMTC Tracker application and stop computing and updating the object position both graphically and internally. Clients can still connect, but they will only receive the last known position of the object.

Implementing a client

The positioning protocol

NOTE: TrackerServer currently only supports one object and its ID is 0.

After that TrackerServer will return to the client the X,Y, and Z position as well as the time in the following format:

<>

The X,Y, and Z components will be integer values that specify the position of the object in the coordinate system used in TrackerServer. The TIME component is the number of seconds elapsed since midnight (00:00:00), January 1, 1970, coordinated universal time, according to the system clock. If an invalid object ID is passed to TrackerServer, it will return "Invalid Object ID" After returning the this information TrackerServer will drop the connection. This component of TrackerServer is multi-threaded will accept multiple simultaneous client connections.

Here is an example Tracker Server client implemented in C++.