Ten Downtown Bars With Patios Fi by pengxiang


									Ten Downtown Bars With Patios
   Final Project for G&G 444
  Created by Kurt Wagner and Jeremy
        Wilz in the Spring of '08
                    Our Mission Statement
•   For our final project, we researched the GPS locations of certain bars in the downtown Madison region. We used
    bars with patios to ensure more accurate GPS readings, as indoor bars did not give us workable data. In addition
    we will display drink specials at these bars and overlay the data into an interactive Flash file that is web-based and
    viewable by all.

•   In researching this topic, we planned to locate GPS locations of these bars, and then in turn, reference them with
    daily drink specials so that people can use the map to find a patio offering drink specials of their choosing.

•   Some key concepts we wanted to address were which bars had patios for seasonal seating, the locations of these
    bars, and their drink special. We also wanted to make sure we included an interactive map so that people
    unfamiliar with the area could locate these bars relatively easily. On this map we will have photos of the bars so
    people will know what to look for in locating the establishments.

•   We used a handheld GPS unit for our measurements. We recorded eight readings over a 15 minute time period.
    From this, we averaged the readings and have a precision between 1 and 10 meters.

•   In planning this project, we hypothsized that our readings were going to be accurate as we were taking eight
    different measurements at each location over a period of 15 minutes. We also theorized that our readings would
    be usable as being outdoors would give us a higher liklihood of attaining more satellite signals to yield more
    accurate results.
                Operational Procedures
•   A: We meet up
•   B: Turn on GPS unit
•   C: Proceed to first location. Find center most open area on each patio. Make sure to have at least
             3 satellites registering, preferably more
•   D: Get waypoint readings over 15 minutes at location
•   E: Record Data
•   F: Proceed to next location
•   G: Repeat steps D and E
•   H: While at each location, collect data on drink specials
•   I: Record Data
•   J: After final location, download data into CSV format onto computer with GPSBabel
•   K: Using GPS Babel, convert CSV into KML format
•   L: Overlay data into GoogleMaps by importing .kml file
•   M: Use this image as base map for Flash file
•   N: Embellish map with interactivity and creative design
•   O: Post map onto World Wide Web for all to see and enjoy
•   P: Presentation of project to class
                        Discussion and Issues
•   One of the issues that we ran into with our readings was that we needed to take readings on multiple days. Our readings
    were taken with the same GPS handheld unit which was a Garmin eTrex Vista Cx with a serial number of A203029360. So
    as far as the hardware goes, the calibrations should have been the same for all of the readings. We could have had
    additional data if we would have used more GPS handheld receivers.
•   Another issue of uncertainty or inaccuracy was the number of satellites accessible for readings. Most of the bars that we
    were measuring had six or seven satellites available which would yield a more accurate result. However, we ran into two
    bars, the Brocach and the Old Fashioned, which only had four and five satellites available, respectively. This was due to the
    location of the bars and the surrounding buildings and obstructions. Since there were fewer satellites available, the reading
    was less precise and thus shows a more sporadic range of latitude and longitude coordinates as can be seen by Figure 1 in
    the back of the packet.
•   Yet another issue of accuracy was dealing with the construction on State St. When measuring State Street Brats, we were
    forced to take our measurements fifteen meters from the actual location we desired. Due to this, our readings were off
    since we could not go into the construction zone. However, the spot in which we took readings, we had six satellites and our
    readings shown seem to be very precise.
•   Another issue faced with uncertainty was when we overlayed the .kml file into Google Maps. We decided to use Google
    Maps instead of Google Earth because our past experience with this class and other Cartography classes have shown us that
    the latitude and longitude points sync up better using a Google Maps mash-up. The concern for uncertainty is still an issue
    for Google Maps, though. This is because we are at Google’s mercy for how accurately they have projected their maps,
    where their standard parallels are located, and how accurate our GPS handheld captured the points. We noticed that the
    points from the overlay are very close to where the actual bars are.
•   The ground truth of actually going to these bars and comparing them to the Google Maps over lay are quite good. The
    points show, with a few exceptions, quite accurately where we took our readings. The Essen Haus in particular shows a very
    high level of precision. Since we didn’t move our GPS receiver and the latitude and longitude coordinates only show a
    variance of 0.00002 decimal degrees, we feel that the ground truth compares very well with our readings.
•   Our overall interpretation of these readings is that they appear to be accurate for the most part. There are a few readings
    that are exceptions to this statement, however. For example, the Memorial Union shows one reading in the Lake. In our
    initial hypothesis, we accounted for one or two erroneous readings. This is why we decided to take multiple readings in the
    same location. By doing this, the overall point which we plotted would be biased to the more precise points.
Map of our GPS Reading Waypoints
•   One thing we learned was to allocate a sufficient amount of time for these types of projects. We initially estimated this
    project to take less time and therefore needed a second day in order to finish up our readings.

•   Another lesson learned was to account for errors with respect to the GPS receiver. There are two issues in particular. One is
    that we could have used an additional receiver to verify the points of our single unit. The other issue is that some of the
    time the GPS receiver would not pick up as many satellites as we would like. To try and counteract that, we found that
    taking multiple readings at one location gave us better precision and we could therefore have more confidence in the points
    that we plotted.

•   A third lesson was to conduct a more thorough evaluation of our test sites. This became an issue to us with the construction
    at State Street Brats. If we had timed our readings for that bar, we could have measured in the area that was occupied by
    construction during the day. This would have been an advantage since the location with construction is where the patio is
    going to be rebuilt.

•   The specifications were met with exception to the readings taken at the Brocach and the Old Fashioned. These points were
    obstructed and could not receive as many satellites as some of the other, more precise points. Comparing our readings and
    our map to the ground truth, we feel that the specifications of one to ten meters of accuracy were fulfilled. We can
    attribute this partly to our initial preperation which included sitting in the middle of the patio so that we were supplied with
    a buffer zone approximately the same size that we aimed for with our accuracy.

•   With the exception of the aforementioned lessons learned as well as some of the issues faced, we would have conducted
    our project quite similarly. Through the course of this project, we gained a larger respect for the preparation time as well as
    the time to conduct the research. We now better understand more of the problems that people encounter in this kind of
    work, both mechanically and socially. We understand that putting more time into these projects and taking more readings
    will yield a better, more accurate, result.
         Basics of Final Map Layout
                                      This is a screen
                                        showing the
                                      functionality of
                                        our map. As
                                      you can see, we
                                       have selected
                                            the Old
                                       Fashioned on
                                      Monday nights.
                                         In turn, the
                                       window in the
                                       middle on the
                                      bottom tells us
                                          the drink
                                        specials, and
                                      the window on
                                      the bottom left
  Here is where Bar                   shows a picture
   images and their                    of the bar and
Lat/Long are displayed                  it’s Lat/Long

To top