=Paper=
{{Paper
|id=Vol-1538/paper-09
|storemode=property
|title=GPS Assistance Taximeter Suited to the Characteristics of a City
|pdfUrl=https://ceur-ws.org/Vol-1538/paper-09.pdf
|volume=Vol-1538
|authors=Julio Santisteban,Ximena Aranzaens Cam
|dblpUrl=https://dblp.org/rec/conf/latincom/SantistebanC15
}}
==GPS Assistance Taximeter Suited to the Characteristics of a City==
https://ceur-ws.org/Vol-1538/paper-09.pdf
7th Latin American Workshop On Communications - 2015
GPS Assistance taximeter suited to the characteristics Of a City
Julio Santisteban Ximena L. Aranzaens
Urb. Campiña Paisajista s/n Barrio Urb. Campiña Paisajista s/n Barrio
de San Lázaro, Arequipa, Perú de San Lázaro, Arequipa, Perú
Universidad Católica San Pablo Universidad Católica San Pablo
jsantisteban@ucsp.edu.pe ximena.aranzaens@ucsp.edu.pe
Abstract— Today almost every one use a taxi, in many It is very common that taximeters use electrical impulses,
countries the rate is agreed upon the service is taken, in many and that they are connected to the car transmission. The
cases the rate is too low or high and the service is not taken, on electrical impulses enable the taximeter to measure the
the other hand taximeters provides good estimated of the service distance and time [15].
but are too complex to install in a cab and maintain. Today we
can get advance of GPS technology to improve and enhance the In this work present a novel taximeter, that consider some
taximeters while the rate is calculate base on the different parameter of geography and socioeconomically situation
features of a city. We propose a novel algorithm to calculate rate from a city.
using GPS data and base on the different features of a city; we
build a prototype and test it with very good results. This work has been developed in a city without any type
of taximeter, and it is a priority improve the service of the
Keywords—gps; taxi; taxímetro public transportation system. Another important information
is that it has a particular geography, so it is important to take
I. INTRODUCTION in mind the different conditions of travel and the different
costs.
T
he GPS is a global positioning system consisting of
an orbit of 24 satellites, which were originally used
for military purposes. Nevertheless, it was opened II. BASIC PARAMETERS
later for commercial and civilian use, [1], [2], [4] and
[6]. The algorithm of a GPS-based taximeter, it is proposed to
provide rate according to the distance travelled, also taking
Consisting on a constellation of 24 satellites that are into account parameters which are:
located in 6 orbital planes with 60º of separation between
each of them [2]. Like any satellite-based system, the timing Place
is very important, that is why they are equipped with 4 atomic
Time
clocks: two cesium and two rubidium.
Fuel Comsuption
The GPS navigation system consists of 3 basic parts, which
are: navigation control center, the terminal and the VHF Aditional Charge
wireless communication network. Vehicle navigation
terminal indicates the current position of the vehicle, its Date (holiday).
speed, and other information in accordance with the GPS Traffic Hour
receiver. After processing this information, it is transmitted
to the control center [4]. Cost of Fuel.
GPS provides 2 service levels which are SPS and PPS [9].
The SPS is the standard obtained with the simple frequency The algorithm to measure distance that has been developed
C/A code. The PPS is the precise service based on code P for is based on different parameters, one of which is the points
dual frequency and is only accessible to authorized users. The that GPS delivers. Hence, the first important thing is to
C/A code is used for civil purposes, L1 and L2 are its consider the frequency of the point taking.
frequencies. A. Definition of the frequency of the point taking
The idea of NMEA is to send a data line, which is To define this, it is necessary to first take measures every
completely independent [11]. All the lines of the NMEA 40 seconds, then every second in moderated traffic
protocol must begin with the symbol "$" [7] and [24], conditions, at an average speed of 40 Km per hour (maximum
followed by the letters GP. The size of each line should not rate allowed) [23].
exceed 80 characters of visible text. There are different types In Figure 1 are the points taken in the first case, the route
of message, and each one of them provides different starts at a ends in b; and you can clearly see that is not feasible
parameters, which gives information. The most basic are the because de reconstructed routes is not real. This is because
GGA, RMC the GSA, [18]. the time between points is too long.
Copyright © 2015 for the individual papers by the papers’ authors. Copying permitted for private and academic purposes. This volume is published and
copyrighted by its editors. Latin American Workshop On Communications' 2015 Arequipa, Peru Published on CEUR-WS: http://ceur-ws.org/Vol-1538/
you can see how would be the first possible division, which
is in concentric circles, the central area of this figure
represents what would be the Centre of the city, and while go
Fig. 1: Case number 1 taken from points every 40 seconds away and exchange areas also would change the base price
given per meter. In this type of graph the entire circumference
will have the same price level and the size of the
circumference will be depending on how many messages are
taken per second of all the GPS receiver module delivery.
And speed of average vehicles have in the city.
The second way is proposed in figure 3 too. Where each
circumference is divided into zones, this makes the method
more specific, but at the same time this brings one greater
complexity, since to make it as similar as possible to the
figure, the streets in the city should be square which
In Figure 2, the second case, where the decision is every unfortunately is not. And also should bear in mind the
second it is viable because you can take the points and azimuth angle of the path that the taxi driver is taking.
reconstruct the real distance that the car is traveling. In this
case the error is less than 1 m that does not change the
measurement considerably. Fig 3: Division into zones concentrically and in each level
Fig. 2: Second case, take points every second
The last form that has been proposed for the calculation of
the rates, as shown in Figure 4, is the divide the city by
. districts and in turn each one in urbanizations. This would be
In addition, it is important to consider that the maximum simpler if all districts were square or at least have regular
speed is 40 Km/h, [29]; this speed may vary depending on the form, but this is not real unfortunately as shown in Figure 5,
place where the system is deployed. It is important that the which is a graph taken from Google Earth, with an example,
space between points will not be so big, for that reason the of the form for a district. This heterogeneous forms makes
next formula is enunciate. impossible to develop this three proposals.
𝑉
𝑡 = 𝑚𝑎𝑥 (1)
11 𝑚 Fig. 4: Division by districts and urbanizations
Where:
DISTRITO
DISTRITO
t: It is the interval to take the points
Vmax: The maximum permitted speed DISTRITO
The system take a point every 11.11 m approximately
DISTRITO
when it is configured to take a point every second, as shown DISTRITO
in Figure 2. Taking this interval of points, we will have 10 División por urbanizaciones
dentro de un distrito
points per block approximately.
Fig. 5: Map of the city with an example of a district
B. Division of the city
To achieve an efficient way of pricing the taxi service, try
different kinds of methods. First dividing the city by sectors,
first urban, semi-urban and rural if they exist, to later divide
the urban sector in areas where the fee will change depending
on some factors, it is important that that the algorithm does
not become too complex division of this sector is as
homogeneous as possible.
At last the algorithm of division chosen is to divide the city TABLE 1: APPROACH GEOGRAPHIC COORDINATES TO METERS
into squares of 1 Km per side, this algorithm has been chosen Coordenadas Geográficas Metros
to simplify the programming, without reducing the system’s
1º 111000 m
efficiency. The division can be seen in Figure 7, several of
the squares divide a single district, and can be differentiated 1’ 18350 m
by percentages that will be multiplied by the rate base. The 1’’ 30.83 m
dividing lines of the quadrants must be parallel to the
meridians and the parallels so that the algorithm works
properly. The three values that were converted to meters are then
added. The same procedure is done for length. After this
process, formula 2 can be used. The distances obtained will
Fig. 6: City divided in squares of 1 Km per side be accumulates and will become the total distance of the
route.
Fig 7: Method to measure distance
III. ALGORITHM
For the calculation of distance, it is important to explain in For the calculation of the fare, in addition to the distance
advance the logic that will used, taking in mind that every travelled, it is fundamental to know the actual quadrant of the
measured point will have its own latitude and longitude. taxi. This will give a percentage that is multiplied to the rate
Knowing that the longitude is parallel to the Equator and the base, which is given by every 20 meters.
latitude is transversal to it. In the worst case, wherever you
go a triangle will be drawn. Otherwise, either the longitude The rate base will be differentiated for each company
or the latitude will increase. according to their expenses, being able to determinate their
rate base for the day. This will be entered in an encrypted way
Taking into account that we will use a point every second, by means of the keyboard’s system. For the implementation
these triangles will be small, as shown in Figure 7, using the of the prototype, was calculated a rate base through the
Pythagorean theorem we can get the distance value. For this different pricing tables that exist in the city, thus, obtaining
the formula we will use for the worst case will be: the rate base of 0.008 as proof.
𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒 = √(𝐹𝑃𝑙𝑎𝑡 − 𝐼𝑃𝑙𝑎𝑡)2 + (𝐹𝑃𝑙𝑜𝑛 − 𝐼𝑃𝑙𝑜𝑛)2 (2) The percentages of each quadrants was calculated based at
empirical study, using the comparative analysis. That was
Where: made with the different tables of prices found at the city. Each
FPlat= Final point in latitude company provides a percentage to the different areas of the
city, this values depends. So this table also need to be
IPlat=Punto Inicial en latitud differentiated for companies. For the implementation of the
FPlon= Final point in longitude prototype the division of the city is shown in table 2 and table
3. The percentages of each quadrants is shown at table 4.
IPlon= Initial point in longitude
TABLE 2: LATITUDE DIVISION IN 1 KM.
To make this possible, it is necessary to convert the
longitude and latitude, which are currently in degrees, Latitude
minutes and seconds; to meters. Taking in consideration that Grades Minutes Seconds Meters
all meridians have an approximate length of 20003,916 km 16° 19 17.82 35695.5906
that go from - 90 to 90 degrees, i.e. 180°: The approximations 16° 19 50.91 36715.7553
of geographical coordinates to meters can be found in table …
1. 16° 28 12.04 52165.5932
16° 28 45.13 53185.7579
TABLE 3: LONGITUDE DIVISION IN 1KM IV. TEST/CASES
Longitude
Grades minutes seconds Meters This section will be an example of follow-up to a possible
71 27 55.06 51642.0998 route for a taxi, and how the rate calculated. It is shown in
71 29 34.48 54707.2184 table 5, in this example, the basic rate taken is 0.008 soles,
… for every 20 meters. The chosen route begins at the plaza de
71 36 47.23 68048.9009 Armas to the Calle Fernandez Davila.
71 37 20.37 69070.6071 In this table you can see that the total at the end of the tour
fare, has a reasonable according to the distance of travel cost,
since this example consists of a short route where the
If the service is during in an hour or zone of traffic, the percentage by area is the same, the final price is a price low.
taximeter sends a high number of frames, if these are more
than 20 each 20 meters, the rate increases in 30% of the rate V. IMPLEMENTATION
base every 20 meters that the car is in the same state. This
amount is determined by the average expenditure which the A. Design
driver uses to move normally, without a situation of high Diagram schematic fact software Proteus you can see in
traffic. Figure 12, this figure shows that to simulate GPS input which
Another situation that also adds a percentage to the rate not is located inside the devices, owning this software; has
base is the day. If the service is being on Sunday or in been used a virtual terminal, which is connected to the port
holidays the basic rates increase in 50% more. Besides if the of the microcontroller PIC 18F4550 serial. [24].
service is taken in hour from 10:00 pm to 4:00 am, the basic Figure 8: Taximeter’s Schematic
rate add a 10% more.
The line ZDA is used to know the date. This line will be
taken into account only at the beginning of the service.
And finally to know if the service is providing to high
hours of the night, used the line called GLL, where gives the
current time in UTC format data.
TABLE 4: PERCENTAGE FOR EACH QUADRANT
Longitude 51642.099 52663.806 53685.5122 54707.2184 55728.9246 56750.6308 57795.7678 58817.474 59840.1051
Latitude
35695.5906 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01
36715.7553 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.05
37735.92 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.05
38756.0847 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.05
39776.2494 0.01 0.01 0.01 0.01 0.03 0.03 0.4 0.001 0.03
40986.9435 0.01 0.01 0.01 0.01 0.03 0.03 0.4 0.4 0.4
42007.1082 0.3 0.3 0.3 0.3 0.3 0.04 0.4 0.4 0.4
43027.2729 0.3 0.3 0.3 0.3 0.3 0.3 0.4 0.4 0.1
44016.2993 0.01 0.02 0.3 0.09 0.09 0.09 0.09 0.4 0.1
45046.0213 0.01 0.02 0.3 0.04 0.05 0.05 0.06 0.4 0.1
46017.7829 0.01 0.02 0.3 0.04 0.04 0.04 0.05 0.06 0.4
47064.7697 0.01 0.01 0.01 0.01 0.01 0.02 0.03 0.4 0.05
48084.9344 0.01 0.01 0.01 0.01 0.01 0.02 0.03 0.4 0.05
49105.0991 0.01 0.01 0.01 0.01 0.01 0.02 0.03 0.4 0.05
50125.2638 0.01 0.01 0.01 0.01 0.01 0.02 0.03 0.4 0.05
51145.4285 0.01 0.01 0.01 0.01 0.01 0.02 0.03 0.4 0.08
Figure 9: Taximeter schematic in Eagle Software In this figure you can see the prototype consists mainly of
an LCD, a GPS receiver, a microcontroller 18F4550 and a
numeric keypad.
VI. FUTURE WORKS
Rate base which is used to calculate a service rate as well
as other values base for the various factors included in the
calculation, should be updated from time to time, some
alternatives is to apply for the driver that you enter these
values but using some sort of encryption or other methods.
Design of the prototype, for convenience in Eagle
schematic diagram was implemented. The design was Or send them through any system, so that it is completely
generated for PCB. transparent to end users as the driver and the passenger. As a
solution for this it is proposed to use a GPRS terminal to send
B. Construction and receive data, and update automatically.
Figure 10: Taximeter with GPS assistance prototype Using this method, there will a large amount of data,
which opens the door to many more utilities and applications
to improve and extend the functionality of the taximeter.
TABLE 5: FOLLOW ROUTE
Origen Origen Destino Destino Distanci Noch Doming Precio Porcentaj Precio Precio
(Nombre (Coordenadas (Nombre (Coordenadas a e o (50%) e por zona anterio total
) geográficas) ) geográficas) (30% r
)
Plaza de Latitud: 16°23'55.32"S Calle Latitud: 16°23'47.18"S 271.54 No No 0.082 Centro 0 0.082
Armas Longitud: Alfonso Longitud: 71°32'7.44"O metros (10%)
71°32'10.91"O Ugarte
Calle Latitud: 16°23'47.18"S Calle Santa Latitud: 495.46 No No 0,14923 Centro 0.081 0.23123
Alfonso Longitud: 71°32'7.44"O Marta 16°23'52.76"S metros 8 (10%) 8
Ugarte Longitud:
71°31'51.78"O
Calle Latitud: Calle Santa Latitud: 16°23'51.68"S 43.40 No No 0,01362 Centro 0.231238 0.24485
Santa 16°23'52.76"S Rosa Longitud: (10%) 8
Marta Longitud: 71°31'50.83"O
71°31'51.78"O
Calle Latitud: 16°23'51.68"S Calle San Latitud: 16°23'53.16"S 63.73 No No 0,01971 Centro 0.244858 0.26457
Santa Longitud: Pedro Longitud: 9 (10%) 7
Rosa 71°31'50.83"O 71°31'49.34"O
Calle San Latitud: 16°23'53.16"S Av. La Paz Latitud: 237.79 No No 0,07193 Centro 0.264577 0.33651
Pedro Longitud: 16°23'59.28"S 7 (10%) 4
71°31'49.34"O Longitud:
71°31'44.40"O
Av. La Paz Latitud: Av. Latitud: 16°24'3.46"S 205.2 No No 0,06216 Centro 0.336514 0.3986
16°23'59.28"S Goyeneche Longitud:71°31'39.03" (10%)
Longitud: O
71°31'44.40"O
Av. Latitud: 16°24'3.46"S Calle Latitud: 16°23'55.86"S 361.74 No No 0,10912 Centro 0.3986 0.5077
Goyeneche Longitud:71°31'39.03" Fernández Longitud: 2 (10%)
O Dávila 71°31'29.65"O
[7] A.Pozo-Ruiz, A.Ribeiro, García-Alegre, M. C., L.García, D.Guinea,
&F.Sandoval. (1999). SISTEMA DE POSICIONAMIENTO GLOBAL
VII. CONCLUSIONS (GPS): DESCRIPCIÓN, ANÁLISIS DE ERRORES,
This work has managed to provide a means of charging, the APLICACIONES Y FUTURO.
system urban transportation taxis, which provide an optimum [8] Lijun, W., &Huanping, M. (2010). The system model and analysis of
all-digital temporal and spatial anti-jamming for GPS receivers.IET
shape of collection where the driver and the user feel satisfied. ConferencePublications, 2010(CP569), 258-261.
GPS technology provides a fully external to this system and [9] Satélites GPS: fundamentos y aplicaciones (2006) Retrieved 28 de
Abril 2012,
not manipulated by the driver, providing some kind of fromhttp://concurso.cnice.mec.es/cnice2006/material121/unidad3/sat_
security to the system, since the data may not be altered by gps.htm
external media. [10] Villar, J., Otero, A., Otero, J., & Sánchez, L. (2009). Taximeter
verification using imprecise data from GPS.Engineering Applications
In addition, the developed algorithm, provides rates of Artificial Intelligence, 22(2), 250-260.doi:
differentiated according to socioeconomic areas to travel, is 10.1016/j.engappai.2008.07.002
given on the taxi route. What regulates the cost in such a way [11] Richard B, L. (1995, Julio). NMEA 0183: A GPS Receiver Interface
that there is a method that is more fair and equitable for the Standard. INNOVATION, 54 - 57.
user, making more accessible to the public in general. [12] Liu, S., Liu, C., Luo, Q., Ni, L. M., &Qu, H. (2011). A visual analytics
system for metropolitan transportation. Paper presented at the
Calculate the taxi fare is not a simple matter must consider Proceedings of the 19th ACM SIGSPATIAL International Conference
on Advances in Geographic Information Systems, Chicago, Illinois.
different factors and each of them must be customized for
[13] Jantarang, S., Potipantong, P., &Worapishet, A. (2002). A low cost real-
each city, have demonstrated that it is possible to have an time intelligent taximeter sensor.APCCAS (1)'02, 217-220.
algorithm that can consider these factors and calculate a rate. [14] Zhang, D., Li, N., Zhou, Z.-H., Chen, C., Sun, L., & Li, S. (2011).
The use of GPS provides data that is essential to replace the iBAT: detecting anomalous taxi trajectories from GPS traces. Paper
odometers, on the other hand must implement an algorithm presented at the Proceedings of the 13th international conference on
that can consider this this type of information to generate an Ubiquitous computing, Beijing, China.
odometer type GPS. [15] Coche, M. d. (2012). ¿Cómo funciona un taxímetro? Retrieved 2 de
Mayo, 2012,fromhttp://www.marcadecoche.com/taximetro.html
The use of GPS also facilitates using a fully external to [16] Casiopea. (2012). Taxis Retrieved 04 de Mayo, 2012,from
automobile technology, since this avoids manipulation of http://wiki.ead.pucv.cl/index.php/Taxi
users, which would avoid that measurement is inaccurate and [17] Corona. (2012). Corona Taxis y RemisesRetrieved 4 de Mayo,
2012,fromhttp://www.taxicorona.com/index.php?ID=c2NvcnBlcnVh
attempt to deceive any user. Y0BzZXJ2aWNpb3M=&MI=c29pY2l2cmVzQGNhdXJlcHJvY3M
To prevent the loss of signal in urban canyons repeaters, [18] SKYLAB. (2012). GPS Module DataSheetDatasheet (pp. 10).
could be installed in such a way that there is signal 99% of [19] CAST, C. F. H. (2012). DISEÑO E IMPLEMENTACIÓN DE UN
PROTOTIPO ELECTRÓNICO PARA EL MONITOREO
travel. This would have to be implemented by the company VEHICULAR PASIVO EN LA CIUDAD DE LATACUNGA,
that traded the finished prototype. UTILIZANDO TECNOLOGÍA GPS. INGENIERO ELECTRÓNICO
EN INSTRUIMENTACIÓN, ESCUELA POLITECNICA DEL
REFERENCES EJÉRCITO EXTENSION LATACUNGA.
[20] Torres, M., Paz, K., & Salazar, F. G. TAMAÑO DE UNA MUESTRA
[1] Walsh, D., Capaccio, S., Lowe, D., Daly, P., Shardlow, P., & Johnston, PARA UNA INVESTIGACIÓN DE MERCADO (pp. 13). Boletín
G. (1997). Real time differential GPS and GLONASS vehicle Electrónico Universidad Rafael Landívar.
positioning in urban areas.Space Comms., 14(4), 203-217.
[21] Labcenter. (2012). labcenter electronics.Proteus PCB Design Packages
[2] Chadil, N., Russameesawang, A., &Keeratiwintakorn, P. (2008, 14-17 Retrieved 28 de Junio, 2012, from
May 2008). Real-time tracking management system using GPS, GPRS http://www.labcenter.com/products/pcb_overview.cfm
and Google earth. Paper presented at the Electrical
Engineering/Electronics, Computer, Telecommunications and [22] Microchip. (2012). Microchip.MPLAB Integrated Development
Information Technology, 2008. ECTI-CON 2008.5th International Environmen Retrieved 28 de Junio, 2012, from
Conference on. http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAG
E&nodeId=140 6&dDocName=en019469&part=SW007002
[3] Karimi, H. A., & Lockhart, J. T. (1993, 12-15 Oct 1993). GPS-based
tracking systems for taxi cab fleet operations. Paper presented at the [23] Ministerio de Transportes, C., Vivienda y Construcción. (2012).
Vehicle Navigation and Information Systems Conference, 1993., REGLAMENTO NACIONAL DE TRANSITO.
Proceedings of the IEEE-IEE. Retrievedfromhttp://www.seguridadidl.org.pe/normas/relevantes/trans
ito.pdf.
[4] Lou, S., Xing, J., Chen, G., Zhang, D., & Li, C. (2008). The Design of
Vehicle GPS Terminal Based on S3C44B0X. Paper presented at the [24] Microchip. (2009). PIC18F2455/2550/4455/4550 Data Sheet 28/40/44-
Proceedings of the 2008 International Conference on Embedded Pin, High-Performance, Enhanced Flash, USB Microcontrollers with
Software and Systems Symposia. nanoWatt Technology (pp. 436). www.microchip.com.
[5] Akopian, D., &Agaian, S. (2005, 17-20 Oct. 2005). A fast position
recovery approach for GPS receivers in jamming environments and a
recommendation for future GPS signal structure. Paper presented at the
Military Communications Conference, 2005. MILCOM 2005.IEEE.
[6] Liang, Z., Amin, M. G., & Lindsey, A. R. (2000, 2000). Subspace
projection techniques for anti-FM jamming GPS receivers. Paper
presented at the StatisticalSignal and Array Processing, 2000.
Proceedings of the Tenth IEEE Workshop on.