A smart driving direction service based on GPS traces using crowd sourcing of Taxi drivers. Find out the fastest driving directions with less online computation according to user inquest. GPSequipped taxis are employed as devices send location information to a centralized sever. A large number of taxi cabs traversing in some areas, for efficient taxi dispatch. Taxis are usually equipped with a GPS sensor, to report their locations to a server at some intervals. A lot of taxis already exist in major areas, generating a huge number of GPS trajectories every day. Taxi drivers are usually found out the fastest route to send passengers to a destination based on their knowledge.
Index Terms |
| GPS trajectory, driving directions,
driving behavior |
INTRODUCTION |
| The traffic problem has analysis here, through the GPS
sensor. The user has to send the inquest to the server
means then it will send through the GPS sensors. Over
crowded has occur in town and cities so the driver can’t
reach the source to destination point in the proper time.
So, through the unceasing viewing the data can be
collected and send through the GPS service. The drivers
can be equipped with some sensor mean we can send the
scattered thinly here and there to view their correct
location, Because of the traffic problem we have lot of
tension and incompetent. |
| Paper [1] describes some applications and algorithms
are used here, real time traffic information is calculated.
A real time monitoring the data can be shown in a curved
line for the source and the destination point. Algorithm
used here is HMM (hidden Marko model), through these
we can easily find the scattered data to identify the
alternative route. |
| Paper [2] describes the overcrowded that occur in the
traffic especially in town and cities. So, the drivers used
to find the alternative way to find the correct path to
reach the destination in the precision time. Sparse data
and probe data are the approaches used sparse data
means scattered thin data can be identified for starting
and the ending point to reach the correct area. Probe data is to analysis the travel time estimation for every vehicle
through the GPS device. |
| Paper [3] describes a shortest path for the alternative
route. Techniques used are pre-compute and proximity,
the calculation of the every route to be changed by the
server and send to the GPS sensor. Then we have to recalculate
the current route for the alternative way to find
the source and destination point. The proximity can be
also a k-candidate path to analysis the k-best alternative
possible way. the quality parameters like cost, response
time, reliability and availability etc. |
| Paper[4] describe a variance-entropy-based clustering
approach used to find out the fastest path. The two-stagerouting
algorithm is used to find the time-dependent
landmark graph for the taxi driver. |
| Paper[5] describe a coherence expanding algorithm to
find the most popular route for the driver. The curved
line only we can find in the coherence expanding
algorithm. To view the source to destination there is no
map is available for the taxi driver. |
LITERATURE SURVEY |
| TITLE: VTrack- Accurate Energy-Aware Road Traffic
Delay Estimation Using Mobile Phones [1] |
| AUTHOR: Arvin Thiagarajan, Lenin Ravindranath |
| CONCEPT: The congestion and traffic analysis are
evaluated in the mobile phone. The overcrowded has
occurred in the origin of incompetent. So the driver got a
lot of tension and wasted fuel. To reduce the traffic
problem viewing the different types of application, GPS,
Wifi, etc. So through these we can avoid the traffic
problem. The GPS can analysis the location and send to
each and every process to calculate travel time. |
| ADVANTAGE: |
| VTrack can use two techniques: |
| 1. Energy consumption |
| 2. Unreliability |
| 1. Energy consumption: The driver has to move
from one location to another location means
there occur some energy consumption loss. For that process the Vtrack provide the less noise
like Wi-Fi sensor application. |
| 2. Unreliability: Some of the phones do not sensor
the data through GPS. So the location can’t be
analysis in proper manner. VTrack provide
HMM model to find the scattered location. |
| DISADVANTAGE: |
| Through the GPS sensor we can’t analysis the
given location properly. The server does not show the
proper location for the taxi drivers. The power can also
be low. |
| TITLE: statistical modeling and analysis of sparse bus
probe data in urban areas. [2] |
| Author: Andrei IU, Richard j. gibbers, David Evans
CONCEPT: Congestion is occurred in town and cities.
Due to this the driver causes more financial loss and
increases the traffic problem so, the data are scattered
here and there (not dense). The traffic can be control by
increasing the capacity of road network. Through the
route we can travel each and every day without the
congestion. The drivers can travel in different way
through the network; we can also increase the capacity.
The drivers can improve their knowledge performance of
these networks. |
| ADVANTAGE: |
| The congestion has been controlled by these techniques |
| 1. Sparse data |
| 2. Probe data |
| 1. Sparse data: the data can be used to analysis the
vehicle sensor through some network. Through these we
have to search the network and send the data to the
driver. |
| 2. Probe data: the probe data is another technique that
can be used for sensor through GPS device. This device
can be attached with the drivers. The probe technique can
be examined thoroughly into the sensor; the data can be
analysis in a sequence of coordinates. The curved line
that follows as it moves through air based on the
techniques. |
| DISADVANTAGE: |
| The AVL data is one of the problem, because it dose not
provide the arrival time and departure time for the driver
properly. So, for that we went for the probe data to solve
the problem. Through the AVL data we cannot update
each and every seconds through GPS. |
| TITLE: A continuous query system for dynamic route
planning [3] |
| AUTHOR: Nirmesh malviya Samuel madden |
| CONCEPT: The unceasing traffic problem has occurred
simultaneously; so we have to find the alternative inquest
path. Then the source and the destination path have been
calculated for daily approach. Here we have to analysis
the techniques as K-path and proximity. Through these
techniques we have to analysis the shortest path. Then
the data has been stored in I phone and I cartel (Which
are capable of being used). The ad-hoc network also be
used but it does not update properly (unceasing) the data.
The ad-hoc network is a plain (temporarily). |
| ADVANTAGE: |
| The unceasing inquest has been used to find out
the powerful way for each driver. The two ways has
been proceed here are |
| 1. Pre-compute |
| 2. Proximity |
| 1. Pre-compute: It has to calculate the alternative
examine way to reach the path. |
| 2. Proximity: It is used to find the nearest way for
drivers. Then we have to analysis through the
graph other than these two techniques we use K
-As-Variance and Y-Moderate. |
| DISADVANTAGE: |
| The algorithm that recalculates the alternative
path as a part of whole route. The update cannot be in a
proper manner; through the weather condition we cannot
identify the path. |
| TITLE: T-drive: driving direction based on the taxi
trajectories. [4] |
| AUTHOR: Jing Yuan, yu zheng, cheng yang zhang |
| CONCEPT: In this paper, GPS-equipped taxies are used
to find out the fastest way for the drivers. The approach
used here are variance-entropy-based clustering to
estimate the travel time. The experienced drivers can’t
find the firm route from source to destination; but the
new driver can’t find the easiest way under the traffic.
So, for the graph we have to use a two-stage routing
algorithm to analysis a easiest way. The time-dependent
landmark graph has been used to find the shortest path
(source-destination). This approach is based on the taxidriver
that is inserted inside the GPS-equipped taxi to
find out the real-time-traffic. |
| ADVANTAGE: This paper is based on drivers intelligent
to find out the source to destination point. A time
dependent landmark graph has been constructing and to
perform a two-stage routing algorithm finding the fastest
path. |
| DISADVANTAGE: The problem is to finding the
shortest path in a real-time-traffic analysis. The updating
can’t be in a frequency manner for the taxi drivers. |
| TITLE: Discovering Popular Routes from Trajectories.
[5] |
| AUTHOR: Zaiben Chen, Heng Tao Shen, Xiao fang
Zhou |
| CONCEPT: The most popular route can be analysis
through the two area (source to the destination) point.
Some of the driver can’t find the easiest way to reach the
end point, for the drivers we have to find the popular way
for each and every driver. To find the easiest way we go
for the algorithm a Coherence Expanding to analysis and
transfer the fastest route for each and every location.
Then we use the Marko chain model is used to drive the
transfer probability for each and every node to transfer
the network. Finally we use the maximum probability
product algorithm to find the fastest route in a breadthfirst
manner. |
| ADVANTAGE: To viewing the best route from one area
to another area, through the coherence algorithm we can
analysis the network for the inexperience drivers to view
from source to destination point. |
| DISADVANTAGE: The view source to destination point
we have to see through the covered line only because
there is no map is available for the drivers. The curved
line only we have to find the most popular route. |
CONCLUSION |
| In this survey paper, we have to formulate the way of
finding the best way for the traffic estimation among the
mobile phone. GPS application is available in the mobile
phone, to analysis the traffic pattern for the drivers. The
entire algorithm is based on the dijkstra’s algorithm. In
future we are going to improve the algorithm to solve the alternative way for the driver, finding the weather
condition. |
References |
- Thiagarajan, L. Ravindranath, K. LaCurts, S. Madden, H.Balakrishnan, S. Toledo, and J. Eriksson, âÃâ¬ÃÅVtrack: accurate, energyawareroad traffic delay estimation using mobile phones,âÃâ¬Ã in Proc.ENSS. ACM, 2009.
- Bejan, R. Gibbens, D. Evans, A. Beresford, J. Bacon, and A.Friday, âÃâ¬ÃÅStatistical modeling and analysis of sparse bus probe data inurban areas,âÃâ¬Ã in Proc. ITS, 2010.
- N. Malviya, S. Madden, and A. Bhattacharya, âÃâ¬ÃÅA continuous querysystem for dynamic planning,âÃâ¬Ã in Proc. ICDE, 2011, pp. 792 âÃâ¬Ãâ803.
- J. Yuan, Y. Zheng, C. Zhang, W. Xie, G. Sun, H. Yan, and X. Xie,âÃâ¬ÃÅT-Drive: Driving Directions Based on Taxi Trajectories,âÃâ¬Ã Proc. 18thSIGSPATIAL IntâÃâ¬Ãâ¢l Conf. Advances in Geographic InformationSystems (GIS), 2010.
- Z. Chen, H.T. Shen, and X. Zhou, âÃâ¬ÃÅDiscovering Popular Routesfrom Trajectories,âÃâ¬Ã Proc. IntâÃâ¬Ãâ¢l Conf. Data Eng. (ICDE), pp. 900-911,2011.
|
Menu