Proposal for the Project: TravelSafe

sunglowmaizeMobile - Wireless

Dec 10, 2013 (3 years and 6 months ago)

101 views

CS6604
-

Spatial Data Management


1



Abstract


This is a proposal for the project
-

TravelSafe, an
end
-
to
-
end location
-
based mobile application for metro
commuters, that facilitates routing between metro stations based
on safety levels of the zones. This application employs two major
dat
asets namely WMATA data and Crime Data. This project is
being carried out as a part of the Course CS6604: Spatial Data
Management. In this paper, we will mainly discuss the
motivation for this project, previous related works, proposed
architecture of our
implementation and tentative software and
hardware requirements.


Index Terms


Project Proposal, Spatial Data Management,
GIS, Android Application, WMATA, Crime Data, Safety
Routing


I.

I
NTRODUCTION

EOGRAPHIC
Information System commonly abbreviated
as GIS
[1], is a system that captures, stores, analyzes,
manages, and presents data that are linked to location(s). In
the simplest terms, GIS is the merging of cartography,
statistical analysis, and database technology. The scope of GIS
extends to various areas
of study such as archeology,
geography, cartography, remote sensing, land surveying,
public utility management, natural resource management,
precision agriculture, urban planning, emergency
management, landscape architecture, navigation, aerial video,
loca
lized search engines and many more. In the recent days
GIS has been increasingly used to provide location
-
based
services such as routing, locating destinations, predicting
traffic and other conditions etc. These services are offered
very commonly in GPS de
vices and mobile phones. Even
though numerous such applications are available to public
there is a constant competition to develop more faster, thinner
and efficient applications that provide solutions to complex
problems.

This project involves designing

and developing such an
application. This application, named Travelsafe, is targeted for
commuters of metro in the Washington DC Metro area. The
main purpose of TravelSafe is to provide the user all the
available routes between a source and a destination,
the time
taken by each route and the ratings of those routes based on
safety levels of the zones that make the route. The safety level
of a zone is determined by the number and type of crimes that
have happened in that zone in the recent times.

We will
be employing the WMATA Data to access
information about metro services and schedules across DC
Metro area. We will also be using the crime data from all the
Sherrif's offices of all the counties in the DC metro area to rate





the different zones. For now we
plan to rate the zones into
three categories namely safe zone, medium zone and
dangerous zone depicted by the colors green, yellow and red
in the display interface respectively. We plan to make this
application user
-
friendly by giving the user the option
to use a
text or a map interface. More technical details follow in the
further sections.

In the next section we discuss some previous related works
and then go into the proposed architecture of the application
and tentative hardware and software requirem
ents. Following
this, we also decide on how to divide the tasks of the project
among the three of us at a high
-
level.


II.

R
ELATED
W
ORK

A lot of research and development has already been achieved
in this area. Research involves developing a mobile
application

framework. Also Thomas Birnkoff suggested
using the geospatial web services for supporting these
applications. Applications have been developed for the user to
get location and access information he needs. Theories and
models for putting security and priv
acy have also been
pursued. Applications to give you a quick access to the local
metro system (Mass Transit [2], iCommute SF[3, 4] and
WMATA) have used location based the spatial location of the
user versus the current schedule of the underlying metro
sy
stem to provide user with convenience of location and time
to travel. Hence applications to address the issues of schedule
and location based schedule have been researched and
developed. Also we have seen applications like Citizen
Connect[4] to collect inf
ormation about the state of the
city/area. Our Application takes these findings and combines
them in a way ensure safe travel routes to the user. Often
shortest routes may even be the fastest, but not necessarily
safest. Combining the two sets of informati
on from the
databases we develop an application that determines the safety
level of the various options of routes one can take from point
A to point B. We will also develop an algorithm to determine
the route that uses safety and time to travel to give an
optimal
safe route.

III.

P
ROPOSED
A
RCHITECTURE


Data


This section gives a high level overview of the input datasets
that will be used in this project. There are mainly two datasets

1.

WMATA data


These are the transparent datasets
provided by Washington Metro A
rea Transit
Authority

G







Proposal for the Project:


TravelSafe

Ashwini Kumble, Reji Panicker, Sindhu Motupalli

CS6604
-

Spatial Data Management


2

2.

Crime data
-

Crime incidents data within DC
metro/northern Virginia areas.


WMATA provides access to its Metro transit data through 14
APIs without any charge that allows developers to create
versatile applications of different platf
orms including mobile
applications. The APIs can be subdivided into two namely
eight methods that describes the rail transit system and
remaining 6 methods that describes the Metrobus activities.
The following table gives a high level overview of each
met
hod provided by WMATA[5]. Each method can be
accessed using either REST or JSON(P) format to make the
development more flexible.


Method

Description

Rail lines

Gives descriptive information about rail
lines like display name (color), Line
code, start a
nd end station codes etc

Rail Stations

Returns list of all stations in the system or
all stations by line

Rail Station
Info

Returns descriptive information about a
single station.

Rail Paths

Returns a list of stations between two
given stations

Rai
l Station
Prediction

Returns train arrival information as it
appears on the Public Information
Displays throughout the system.

Rail incidents

Returns rail incidents as they appear on
the the Public Information Displays
throughout the transit system.

El
evator
incidents

Returns elevator and escalator statuses as
they appear on the Public Information
Displays throughout the transit system

Station
Entrances

Returns entrances (including elevators) to
Metro stations. Entrances are ordered by
distance from l
atitude/ longitude, if
provided

Bus Routes

Returns a list of all bus routes

Bus Stops

Returns the list of all bus stops . Stops are
ordered by distance from latitude/
longitude, if provided.

Bus Schedule
by Route

Returns the bus schedule associated w
ith
a requested route

Bus Route
Details

Returns a sequence of lat/long points
which can be used to describe a specific
bus route

Bus Positions

Returns the real
-
time positions of each
bus travel a specified route inside
specified area. Bus position info
rmation
is updated every two minutes or less

Bus Schedule
by Stop

Returns the bus schedule for a specific
bus stop




Crime incidents data is provided online by Metropolitan
Police Department (MPD) through DC Catalogs.
The dataset
contains a subset of
locations and attributes of incidents
reported in the Analytical Service Application. The attributes
of this dataset includes crime (name), reported time, offense
like robbery or theft or arson, the method used for the crime,
brief narrative, location deta
ils etc. The detailed explanation of
different fields within the dataset is given in Data Catalogue
maintained by DC[6]



System Architecture


Fig1: System Architecture




Application would have a client server model with a thick
client running on top of
Android Application Framework(top
level of android architecture). The server is run on cloud
platform(or in an Application server hosted on an independent
system). Server hosts the crime database. Both client and
server has access to WMATA data through th
e public APIs.


The server periodically performs data mining operation on
crime data with respect to the metro station and bus service
areas. Based on the result, the stations and service areas would
be flagged with different safety color codes. The serve
r
provides these results as APIs. Depending upon the size of the
data being passed we are planning to implement these services
using either JSON or Google's Protocol Buffers. Proper
indexing of crime data is required to increase the data mining
operatio
ns within the



Android application(client) can be divided into two tiers. The
user interface layer and the Logic layer. The user interface
layer allows application users to get current location or enter a
specific address as the source and then enter a d
estination
address. The logic layer gets the relevant route information
from WMATA API which includes both rail and bus routes
and corresponding safety zoning informing from the server.
With these data, the application shows the safest route to the
Parsing

Routing

Priority

Spatial

Data

Mining

User Interface

Android
Application

Framework

WMATA

A
P
I

Cri
m
e

Data












Client












Server

CS6604
-

Spatial Data Management


3

user w
ith appropriate color codes indicating the safety zones
throughout the route. If multiple routes are available, the
application would display the most safest route with an option
for the user to select other routes that are less safe. For this we
are plann
ing to overlay metro map on top of google maps
layer. User will also have an option to get the routes in a text
format.


IV.

D
EVELOP
T
OOLS


For the server, we plan to use Amazon S3 for scalability and
efficiency. In an event we cannot use S3, we may also use
the
JBoss Application server, to host the data mining code and the
tables which contain the data pertaining to the application. We
may use Postgresql or Spatialite to store our complex
relational database. On the client side we will build a simple
API, usi
ng Java (SDK 6). We will also design Junit test cases
to test the process.






R
EFERENCES


[1]

GIS
. Available:
http://en.wikipedia.org/wiki/Geographic_information_system

[2]

Mass
Transit
. Available:
http://www.sparkfishcreative.com/

[3]

iCommuteSF
. Available:
http://www.jazzmoonstudio.com/http:/apptight.
com/ICommut
eSF.aspx

[4]

Citizen Connect
. Available:
http://www.cityofboston.gov/doit/apps/citizensconnect.asp

[5]

WMATA
Available:
htt
p://developer.wmata.com/docs

[6]

Crime Data
. Available:
ttp://data.dc.gov/Metadata.aspx?id=3


A
PPENDIX

Task Division


Client Interface Layer


Ashwini and Reji

Client Logic Layer


Ashwini and Reji

Server Data and Data Mining Layer


Sindhu


Schedule


Fe
bruary 28
th



Detail Design of Data Structure at Server and
Data Mining Technique

March 7
th



Detail Design of Client Layers

March 14
th



Begin Server Setup

March 21
st



Begin Implementation