Intelligent Vehicle System in Undefined EnvironmentIntelligent Vehicle System in Undefined Environment
This paper discusses the issue of detection of collision of a four-wheel vehicle on road in an undefined environment that is road traffic in small towns and cities. Initially the system is implemented to avoid the border hits of the vehicles during night time because of the inability of driver’s vis...
Saved in:
| Published in: | Электронное моделирование |
|---|---|
| Date: | 2008 |
| Main Authors: | , |
| Format: | Article |
| Language: | English |
| Published: |
Інститут проблем моделювання в енергетиці ім. Г.Є. Пухова НАН України
2008
|
| Subjects: | |
| Online Access: | https://nasplib.isofts.kiev.ua/handle/123456789/101564 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Journal Title: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Cite this: | Intelligent Vehicle System in Undefined Environment / R.V. Krishnaiah, K. Soundara Rajan // Электронное моделирование. — 2008. — Т. 30, № 2. — С. 93-108. — Бібліогр.: 19 назв. — англ. |
Institution
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
nasplib_isofts_kiev_ua-123456789-101564 |
|---|---|
| record_format |
dspace |
| spelling |
Krishnaiah, R.V. Soundara Rajan, K. 2016-06-05T08:42:41Z 2016-06-05T08:42:41Z 2008 Intelligent Vehicle System in Undefined Environment / R.V. Krishnaiah, K. Soundara Rajan // Электронное моделирование. — 2008. — Т. 30, № 2. — С. 93-108. — Бібліогр.: 19 назв. — англ. 0204-3572 https://nasplib.isofts.kiev.ua/handle/123456789/101564 This paper discusses the issue of detection of collision of a four-wheel vehicle on road in an undefined environment that is road traffic in small towns and cities. Initially the system is implemented to avoid the border hits of the vehicles during night time because of the inability of driver’s visual capacity. This system is extended to find the collision point in an undefined that is unknown environment where the approach of the vehicle is unrestricted. This system directs the driver, based on collision point information. The advisory information helps the driver to avoid collision. This system need not be installed in all the vehicles in the environment. It needs to be installed in the vehicle that requires advisory information to the driver. Рассмотрена задача распознавания ситуации с четырехколесным транспортным средством на дороге в условиях неопределенности окружающей обстановки, например при прохождении трафика в поселке и городе. Первоначально система использовалась для предотвращения выхода транспортного средства за границы дороги в ночное время вследствие ограниченности визуального обзора для водителя. Система предлагается в качестве источника консультативной информации для водителя транспортного средства. Розглянуто задачу розпізнавання ситуації з чотириколісним транспортним засобом на дорозі в умовах невизначеності навколишнього стану, наприклад при проходженні трафіка у селищі та місті. Спочатку систему використовували для запобігання виходу транспортного засобу за межи дороги вночі внаслідок обмеженості візуального огляду для водія. Систему запропоновано як джерело консультативної інформації для водія транспортного засобу. en Інститут проблем моделювання в енергетиці ім. Г.Є. Пухова НАН України Электронное моделирование Применение методов и средств моделирования Intelligent Vehicle System in Undefined EnvironmentIntelligent Vehicle System in Undefined Environment Article published earlier |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| title |
Intelligent Vehicle System in Undefined EnvironmentIntelligent Vehicle System in Undefined Environment |
| spellingShingle |
Intelligent Vehicle System in Undefined EnvironmentIntelligent Vehicle System in Undefined Environment Krishnaiah, R.V. Soundara Rajan, K. Применение методов и средств моделирования |
| title_short |
Intelligent Vehicle System in Undefined EnvironmentIntelligent Vehicle System in Undefined Environment |
| title_full |
Intelligent Vehicle System in Undefined EnvironmentIntelligent Vehicle System in Undefined Environment |
| title_fullStr |
Intelligent Vehicle System in Undefined EnvironmentIntelligent Vehicle System in Undefined Environment |
| title_full_unstemmed |
Intelligent Vehicle System in Undefined EnvironmentIntelligent Vehicle System in Undefined Environment |
| title_sort |
intelligent vehicle system in undefined environmentintelligent vehicle system in undefined environment |
| author |
Krishnaiah, R.V. Soundara Rajan, K. |
| author_facet |
Krishnaiah, R.V. Soundara Rajan, K. |
| topic |
Применение методов и средств моделирования |
| topic_facet |
Применение методов и средств моделирования |
| publishDate |
2008 |
| language |
English |
| container_title |
Электронное моделирование |
| publisher |
Інститут проблем моделювання в енергетиці ім. Г.Є. Пухова НАН України |
| format |
Article |
| description |
This paper discusses the issue of detection of collision of a four-wheel vehicle on road in an undefined environment that is road traffic in small towns and cities. Initially the system is implemented to avoid the border hits of the vehicles during night time because of the inability of driver’s visual capacity. This system is extended to find the collision point in an undefined that is unknown environment where the approach of the vehicle is unrestricted. This system directs the driver, based on collision point information. The advisory information helps the driver to avoid collision. This system need not be installed in all the vehicles in the environment. It needs to be installed in the vehicle that requires advisory information to the driver.
Рассмотрена задача распознавания ситуации с четырехколесным транспортным средством на дороге в условиях неопределенности окружающей обстановки, например при прохождении трафика в поселке и городе. Первоначально система использовалась для предотвращения выхода транспортного средства за границы дороги в ночное время вследствие ограниченности визуального обзора для водителя. Система предлагается в качестве источника консультативной информации для водителя транспортного средства.
Розглянуто задачу розпізнавання ситуації з чотириколісним транспортним засобом на дорозі в умовах невизначеності навколишнього стану, наприклад при проходженні трафіка у селищі та місті. Спочатку систему використовували для запобігання виходу транспортного засобу за межи дороги вночі внаслідок обмеженості візуального огляду для водія. Систему запропоновано як джерело консультативної інформації для водія транспортного засобу.
|
| issn |
0204-3572 |
| url |
https://nasplib.isofts.kiev.ua/handle/123456789/101564 |
| citation_txt |
Intelligent Vehicle System in Undefined Environment / R.V. Krishnaiah, K. Soundara Rajan // Электронное моделирование. — 2008. — Т. 30, № 2. — С. 93-108. — Бібліогр.: 19 назв. — англ. |
| work_keys_str_mv |
AT krishnaiahrv intelligentvehiclesysteminundefinedenvironmentintelligentvehiclesysteminundefinedenvironment AT soundararajank intelligentvehiclesysteminundefinedenvironmentintelligentvehiclesysteminundefinedenvironment |
| first_indexed |
2025-11-25T20:43:21Z |
| last_indexed |
2025-11-25T20:43:21Z |
| _version_ |
1850530625043300352 |
| fulltext |
R. V. Krishnaiah
Department of Computer Science
Engineering Chirala Engineering College
(Chirala, Andhra Pradesh, India, E-mail: vkrishnaiahr@yahoo.com),
K. Soundara Rajan
JNTU college of Engineering
(Ananthapur, Andhra Pradesh, India, E-mail: principalj ntucea@yahoo.co.in)
Intelligent Vehicle System in Undefined Environment
(Recommended by Prof. H. Rao)
This paper discusses the issue of detection of collision of a four-wheel vehicle on road in an unde-
fined environment that is road traffic in small towns and cities. Initially the system is imple-
mented to avoid the border hits of the vehicles during night time because of the inability of
driver’s visual capacity. This system is extended to find the collision point in an undefined that is
unknown environment where the approach of the vehicle is unrestricted. This system directs the
driver, based on collision point information. The advisory information helps the driver to avoid
collision. This system need not be installed in all the vehicles in the environment. It needs to be
installed in the vehicle that requires advisory information to the driver.
Ðàññìîòðåíà çàäà÷à ðàñïîçíàâàíèÿ ñèòóàöèè ñ ÷åòûðåõêîëåñíûì òðàíñïîðòíûì ñðåäñòâîì íà
äîðîãå â óñëîâèÿõ íåîïðåäåëåííîñòè îêðóæàþùåé îáñòàíîâêè, íàïðèìåð ïðè ïðîõîæäåíèè
òðàôèêà â ïîñåëêå è ãîðîäå. Ïåðâîíà÷àëüíî ñèñòåìà èñïîëüçîâàëàñü äëÿ ïðåäîòâðàùåíèÿ
âûõîäà òðàíñïîðòíîãî ñðåäñòâà çà ãðàíèöû äîðîãè â íî÷íîå âðåìÿ âñëåäñòâèå îãðàíè÷åí-
íîñòè âèçóàëüíîãî îáçîðà äëÿ âîäèòåëÿ. Ñèñòåìà ïðåäëàãàåòñÿ â êà÷åñòâå èñòî÷íèêà êîí-
ñóëüòàòèâíîé èíôîðìàöèè äëÿ âîäèòåëÿ òðàíñïîðòíîãî ñðåäñòâà.
K e y w o r d s : The keywords are: undefined path, collision detection, automatic dipper, colli-
sion warning, collision avoidance.
1. Introduction. Intelligent techniques as a part of decision-making can be very
effective. Fay [1] has developed a dispatching support system for use in railway
operation control systems, which requires expert knowledge in fuzzy rules of the
IF-THEN type. The concept «intelligence» seems to be very popular among the
transport management systems. In these cases the intelligence usually means
systems capability to adapt when surrounding conditions change or its capability
to learn from data. Continuously increasing amount of vehicles in road, air and
marine transport areas has put the pressure on the research of these problem ar-
eas. More economical, more efficient and thus more intelligent methods have to
be developed to deal with these challenging problems.
ISSN 0204–3572. Ýëåêòðîí. ìîäåëèðîâàíèå. 2008. Ò. 30. ¹ 2 93
�������������
�
�
��
���
���
�����
�����
This paper presents a novel technique, which is utilized to identify the mo-
bile vehicle characteristics and used to define the path of the vehicles with the
help of system processed collision detection point. The threat severity is also dis-
cussed. The basic system is designed and implemented with discrete electronic
components with minimal functionality of avoiding the border hits or colli-
sions of the opposite coming vehicles due to the inability of light perception by
the driver. This product is tested and verified with good number of autonomous
vehicles traveling in non-defined paths.
The concept used in the above basic system is to work with the on-off de-
vices based on the light intensity, which eliminates the manual operation. This
method is meant for the non-specific path travel of the four-wheelers during
nighttime. The threat probability of different orientations is discussed. The dis-
crete components used are optical sensors, solid-state relays. This product is de-
signed for Indian environment of four wheelers travel to avoid the accidents be-
cause of considerable reasons.
The discrete system explains the implementation using discrete components
such as transistors. The alternative is the micro controller based system for the
same purpose, which is extensively used, in data acquisition and processing, to
control the signals acquired from different vehicles in the unknown environ-
ment. The data acquisition and processing describes how to collect the data re-
quired for calculating the collision point. The data acquisition and processing
uses the concept of multiplexing while processing the acquired data and takes
the decisions to communicate to the advisory system. The fourth part describes
the mathematical model for simulating the environment and finds the parameters
of the collision point. In this part three cases are categorized and formulae in re-
spect of each case are derived.
The first and more important, reason for automating the driving process is
safety. In 2001, there were approximately 6,393,000 automotive accidents, lead-
ing to over 41,000 deaths and more than 3,000,000 injured [2] in US as per de-
partment of transportation. The causes of these accidents come from a wide ar-
ray of backgrounds, but all lead to being error on the part of the driver. If this er-
ror has been removed, even in the highway setting, the number of accidents each
year would surely decrease. In 1945, Teetor, inspired by the poor driving habits
of his lawyer friend, invented cruise control [3]. With the invention of cruise
control, the burden of driving was decreased. With a charge as ambiguous as
this, the focus in automation has been spread in many different directions. There
are those who focus on control strategies in adaptive cruise control [4], whose
focus is to slow a vehicle as it approaches a slower-moving vehicle ahead.
Why automate the driving task? One of the major reasons is safety. In 2000,
there were approximately 6,394,000 police reported motor vehicle traffic cra-
R. V. Krishnaiah, K. Soundara Rajan
94 ISSN 0204–3572. Electronic Modeling. 2008. V. 30. ¹ 2
shes, resulting in 3,189,000 people being injured and 41,821 lives lost [5] in US
as per department of transportation. Accidents on our roadways not only cause
injuries and fatalities, but they also have a huge economic impact [6]. According
to [7], driver distraction was a factor in 11% of fatal crashes and 25—30% of in-
jury and property damage- only crashes in 1999.
Traditionally two basic approaches are being used for path generation of an
autonomous mobile vehicle, fixing the total path in a known environment, and
instantaneous fixing of path directions depending on situations in an unknown
environment [8]. An intelligent vehicle may be defined as a vehicle that senses
the environment and produces some automatic action or driver advisory. Vehi-
cle recognition plays an important role in the field of road traffic monitoring and
the collision avoidance purpose. The problems are of different types such as ve-
hicle location identification, vehicle behavior identification. Conventional loop
detectors and video based systems are found to be quiet inefficient in solving
these problems.
Harlow et al [9] have proved that cameras based upon range sensors are not
sensitive to lightning and other environmental conditions. Goerick et al [10]
have applied neural networks for vehicle detection called car track system in
which monocular visual sensor system used, to identify rear frontal views of au-
tomobiles and image sequences taken from the viewpoints of a following car. A
method for vehicle classification using models and neural networks is based on
3D structured model [11]. Acoustic signature of traveling vehicles can be used
for vehicle classification [12]. Pasquerillo et al [13] have proposed the automatic
target recognition system for naval traffic control using neural networks. A com-
prehensive study of road safety (Treat et al 1977) found that human errors were
the sole cause in 57% of all accidents. Only 2.4% were due to mechanical fault
and 4.7% were caused by environmental factors.
The accidents are occurring due to the following reasons during nighttime.
They are identifying the border of the opposite coming vehicle, sleeping of the
driver, high intensity of the headlights, which are not operating properly. The
discrete system of this paper gives the solution for the first and third reasons. The
driver is not interested to use one hand for switch on and off the headlights dur-
ing the vehicle coming in opposite direction. Because of this reason the driver
can’t estimate the border of the vehicle coming in opposite direction and thus
collision may occur at their edges in most of the cases. This type of collision can
be avoided by using the system, which is presented in this paper explained as
discrete system.
2. Discrete system. The discrete form of the collision avoidance system is
designed to avoid accidents during night time. This system includes a light-de-
tecting resistor whose resistance varies with light intensity. It will operate for all
Intelligent Vehicle System in Undefined Environment
ISSN 0204–3572. Ýëåêòðîí. ìîäåëèðîâàíèå. 2008. Ò. 30. ¹ 2 95
types of head light intensities. The Light sensor is connected in a bias circuit so
that, based on the light the transistor is driven into ON and OFF state, which will
drive the headlights. The relay is used between the system and the headlight. Ini-
tially mechanical relays are used which require more response time. Later solid
state relays are used for better response as well as better rise time. The system is
designed for a cycle time of 1 minute.
This system is fabricated and tested on many number of vehicles, which
travel on the road during night time for a considerable length of time. The motor
vehicle inspector and the owners of the vehicles are satisfied with this experi-
ment. They felt that it is advantage because driver need not put the efforts for on
and off the light manually and the synchronization may not be achieved with
manual system.
The block diagram of the system described in this part is shown in Fig. 1.
The Light sensor converts the physical quantity that is Light into its electrical
quantity like resistance. The transistor biasing circuit converts the small varia-
tions of the Light sensor output to the desired level by acting as an amplifier in
active region. The transistor stage is used for on and off purpose with saturation
and cutoff regions. The comparator stage contains two input signals. One signal
is obtained from the sensor biasing stage. The second signal is the reference sig-
nal, which can be varied with the help of potentiometer adjustment such that the
driver will adjust the reference based on his vision capability.
The output of the comparator is thus dependent on the adjustment to the in-
put made by the driver. This stage uses IC741.We design pulse generator stage
to our time period and level of signal. The best-suited generator for this stage is
A-stable multivibrator. This stage uses IC555. In the normal case the pulse gen-
erator is off and the headlights are continuously on. Only when the light intensity
of the on coming vehicle is sensed the system starts blinking the head lights. The
pulses generated from the pulse generator stage must directly drive the switch to
operate the headlight, if not we introduce the driving circuit to drive the switch.
The switches are either electromechanical or solid state relays. The later one is
preferred in this paper.
R. V. Krishnaiah, K. Soundara Rajan
96 ISSN 0204–3572. Electronic Modeling. 2008. V. 30. ¹ 2
Transistor Biasing
Circuit
Pulse Generator Switch & Driving Head Light
Photo Sensor Comparator Stage
Fig. 1. Automatic head light control system
3. Micro controller based system. The inventions of the integrated circuit
(IC) and later, the microcomputer, have been the major factors in the develop-
ment of electronic control in automobiles. The system described in discrete sys-
tem consists of solid-state discrete devices with which it is difficult to process
and handle large data. Hence we have developed a technique using the principle
of data acquisition to acquire data and process it using a micro controller. In this
technique the first two stages of the part1 that is in Fig.1 are same and the later
stages are different.
The next stage in this technique is the analog to digital converter (ADC),
which converts the analog input into digital output. This output is given to the
micro controller with the help of interfacing devices. The driver sets the refer-
ence value related to vision capability. The driver or user enters this reference in-
put through programming the micro controller.
The micro controller generates the square wave if the light intensity of the
opposite coming vehicle falls on the sensor. The headlights are operating on and
off synchronously with the opposite vehicle headlights. At any time only
one-vehicle headlights are on if two vehicles travel in opposite direction. This
technique is advantageous than the first one. Although the micro controller as
used here is not really necessary, it is introduced as an evolutionary step for the
third stage. This technique is applied in the data acquisition and processing of the
paper, which describes the methodology of collision avoidance for vehicles trav-
eling in undefined path. The block diagram of the second technique described in
part 2, is shown in Fig.2. The data acquisition and processing of this paper ex-
plains how the vehicle or object is identified, its position, velocity and the ex-
pected collision point and collision time.
4. Data acquisition and processing. This part is meant for identifying the
moving objects surrounding the moving vehicle on the road. The collision point and
time are calculated by the micro controller and communicated properly to the driver
by advisory system. As a further step of safety due to non-attention of the driver this
system can control the vehicle, which will be doing as a later stage.
Intelligent Vehicle System in Undefined Environment
ISSN 0204–3572. Ýëåêòðîí. ìîäåëèðîâàíèå. 2008. Ò. 30. ¹ 2 97
Transistor Biasing
Circuit
Controller Inerface device Head Light
Photo Sensor ADC interface
Fig. 2. Automatic head light controller system
The existing systems available in cars are adaptive cruise control, in which
the car slows if it detects a slower moving vehicle in front of it, is starting to be-
come available on higher-end models. In addition, some cars come equipped
with sensors to determine if an obstacle is near and sounds an audible warning to
the driver when it is too close [14]. The availability of this feature costs high.
With lateral control technique, the driver would be able to remove his hands
from the steering wheel and let the car steer itself. Here, the idea is that the car
has some desired path to follow. Sensors on the car must be able to detect the lo-
cation of the desired path. The error between the desired path and the car is cal-
culated and the microcomputer acting as the controller determines how to turn
the steering wheels to follow the correct path. The lateral controller’s purpose is
to follow the desired path. It does not determine what the desired path is. A
higher-level planner is responsible for that task. It only needs to know the car’s
location with respect to the desired path [14].
This paper describes a common system, which can track the vehicles in un-
known environment.
The schematic diagram of this technique is shown in Fig 3. In the first stage
the sensors are placed in the four corners of the vehicle in all directions. Those
sensors are rotated between +45 and –45 degrees with the help of stepper motor
at an appropriate speed. The rotation of four sensors covers all the area surround-
ing the vehicle. The stepper motors are driven by using a micro controller, which
knows the angle of rotation at any instant thus the angle of the moving object, is
known surrounding the vehicle in any direction. Based on the delay between
emission and collection of rays, the micro controller calculates the distance of
the moving object surrounding the vehicle in any direction. Hence with the
above two principles we are calculating the distance and angle of the object. By
knowing this data the collision point and time are also calculated.
The outputs of the sensors are connected to individual analog to digital con-
verters. The digital outputs are multiplexed and given to the micro controller
through interface device. The output of the micro controller is given to the advi-
sory system, which is guiding the driver to avoid collision.
The selection ADC is based on the application and the accuracy required.
The micro controller processes these data. The future extension of this project
deals with the vehicle control, which includes the technology of mechanical
parts and is described in literature [15—19]. The central idea on the deci-
sion-making is that the moving objects are recognized as harmful or as harmless.
Harmful objects are those, which approach towards vehicle and harmless ones
move away from the vehicle. The loci of harmful objects are evaluated in Fig 4.
The micro controller uses the level of threat also while finalizing the decision to
the advisory system. The threat level the cellular logic rule for fixation of path to
R. V. Krishnaiah, K. Soundara Rajan
98 ISSN 0204–3572. Electronic Modeling. 2008. V. 30. ¹ 2
avoid the collision is shown in Fig. 3. and explained the rules of threat in next
section. In Fig. 3 the position of the object in future and present is shown with re-
ducing the threat rate, that is object is moving from high threat region to low
threat region. The intended path of the object is given by R � L0. Normal speed
Intelligent Vehicle System in Undefined Environment
ISSN 0204–3572. Ýëåêòðîí. ìîäåëèðîâàíèå. 2008. Ò. 30. ¹ 2 99
Sensor 1 ADC
Sensor 4 ADC
M
U
L
T
I
P
L
E
X
E
R
C
O
N
T
R
O
L
L
E
R
Advisory system
SM interface
SM interface
Fig. 3. Block diagram of collision avoidance system in undefined path
High level threat
Present position of object
H1
H7
H2
H6
H3
H5
H8 H4R
Low level threat
Future position of object
L1
L7
L2
L6
L3
L5
L8 L4L0
No threat region
Present position of object Direction of motion of the object
Fig. 4. Threat calculation
of object is denoted as S. A path is fixed by the object depending on the positions
and speeds of obstacles. At a time instant the object dynamics is determined by
the ordered. pair < S, R � Lj>.
There are 8 loci corresponding to less harmful objects approaching the ob-
ject in directions: Lj R� ,1 8� �j . So R Lj� is a function of Lj R� , and S is a
function of Hk R� ,1 8� �k . Now the cellular logic rule that fixes the instanta-
neous path is:
� � �� � � � � � �S R Lj R Lj j Hk R k, ( , , , )� 1 8 1 8 .
5. Simulation. This part gives the mathematical model of the system which
takes the distance of the vehicle, angle of arrival and the velocity as input and de-
rives the following outputs which are representing the collision point. They are
the distances from all vehicles in an unknown environment to the collision point,
the angle of collision and the velocities of those vehicles leads to find the first
collision span time. In this paper three different cases are considered. The first
case is the vehicles are approaching approximately with 0 degrees that is range
of –2 to +2 degrees. The second case is the vehicles are approaching approxi-
mately with 180 degrees that is range of 178 to 182 degrees. The third case is that
vehicles are approaching with other range of values. These cases are described in
detail later.
Case 1. This is the case in which the vehicles travelling in opposite direc-
tion. Diagrammatic representation is given below. The input values provided by
sensors such as
distance of other vehicle from our vehicle (X);
velocity of our vehicle V1;
angle with which the other vehicle is approaching that is 0 degrees in this
case.
The assumption is that vehicle collides at some point which is at a distance
X1, X2 from two vehicles A and B respectively.
Suppose vehicles collide at point Ñ. And distances from A, B are X1, X2 re-
spectively. From above assumptions A and collides after certain time t, with
their respective distances. (Fig. 5). For vehicle A ,
t X V�
1 1
/ (1)
R. V. Krishnaiah, K. Soundara Rajan
100 ISSN 0204–3572. Electronic Modeling. 2008. V. 30. ¹ 2
A
C
B
X
X
1
X
2
Fig. 5. Vehicles travelling in opposite direction
For vehicle B,
t X V�
2 2
/ . (2)
From Fig. 5 we have
X X X
2 1
� � . (3)
Substitute 3 in 2 and by equaling 1 and 2 we get,
X
V
X X
V
1
1
1
2
�
�
, X
X V
V V
1
1
1 2
�
.
(4)
By substituting 4 in 3 we get,
X
X V
V V
2
2
1 2
�
.
(5)
From equations (4) and (5) we have X1 and X2 respective distances from ve-
hicles A and B. We have angle as 0°.
Case 2. This is the case when vehicles coming from back side of our vehicle.
Diagrammatic representation is given below. The input values provided by sen-
sors such as
distance of other vehicle from our vehicle(X);
velocity of our vehicle V1;
angle by which other vehicle is approaching, in this case that is 180°.
The assumption is that vehicle collides at some point which is at a distance
X1, X2 from A and B respectively. Suppose vehicles collide at point C. And dis-
tances from A, B are X1, X2 respectively. From above assumptions A and B col-
lides after certain time t, with there respective distances (Fig. 6). For vehicle A,
t X V�
1 1
/ (6)
For vehicle B ,
t X V�
2 2
/ . (7)
From Fig. 6 we have
X X X
2 1
� . (8)
Substitute (8) in (7) and by equaling (6) and (7) we get,
X
V
X X
V
1
1
1
2
�
, X X
V
V V
1
1
2 1
�
�
�
�
�
� .
(9)
By substituting (9) in (8) we get,
X X
V
V V
2
2
2 1
�
�
�
�
�
� .
(10)
Intelligent Vehicle System in Undefined Environment
ISSN 0204–3572. Ýëåêòðîí. ìîäåëèðîâàíèå. 2008. Ò. 30. ¹ 2 101
Case 3. This is the case that the vehicle coming towards our vehicle, such
that it makes some angle with our vehicle. The input values provided by sensors
such as
distance of other vehicle from our vehicle (X);
angle made by that vehicle with our vehicle (�);
velocity of our vehicle V1.
The assumption is that vehicle collides at some point with an angle at colli-
sion point is �. We derive from below diagram. X2 is the distance of collision
point from the approaching vehicle, X1 is the distance of collision point from our
vehicle that is where system is installed (Fig. 7). For vehicle A, t X V�
1 1
/ . For
vehicle B, t X V�
2 2
/ . By equalizing both equations of vehicle A and B we have,
X
V
X
V
1
1
2
2
� , X
V
V
X
1
1
2
2
� .
(11)
From base of triangle AOC we have
x x xcos cos� � �
1 2
; x
x x
2
1
�
�cos
cos
�
�
.
(12)
R. V. Krishnaiah, K. Soundara Rajan
102 ISSN 0204–3572. Electronic Modeling. 2008. V. 30. ¹ 2
A CB
X
X
1
X
2
Fig. 6. Vehicles following in the same direction
A
C
90°
B
X
X cos �2
X cos Ô
OX
1
X
2
Ô �
Fig. 7. Vehicles approaching with an angle
From the triangle AOB ( Fig. 8). The side x1 and angle Ô are known, and hence
other side x2 is,
x x x x x
2
2
1
2 2
1
2� � cos�. (13)
From the triangle ABN we have
AB2
=BN 2
+ AN 2
;
x x x x x2
2 1 1
2
1
2 2
� ( cos ( cos�� ��
�
. (14)
Thus we are having four equations and four unknowns, by solving these
equations we get
x
x
1
2
2
3� � �
cos
( cos cos cos
�
� � �� � ��.
. (15)
We consider the value of X1 as Z. By substituting Z value in X2 we get:
x z x xz
2
2 2
2� � cos�. (16)
We consider value of X2 as p. By substituting values of X1, X2 in V2, cos �
we get:
V
V
z
p
2
1
�
(17)
cos
cos
�
�
�
�x z
p
.
(18)
From equations (5)—(8) we get x1, x2, v2, cos �.
Intelligent Vehicle System in Undefined Environment
ISSN 0204–3572. Ýëåêòðîí. ìîäåëèðîâàíèå. 2008. Ò. 30. ¹ 2 103
A
N
M
C
90°
90°
B
O
Ô �
�
Fig. 8. Calculation of angle of collision
6. Algorithms. The algorithms show the procedure how to acquire the data
from sensor and processed it. A1 and A2 are the minimum and maximum angle
with which sensor is rotating. A1 and A2 depend on the number of sensors used.
A l g o r i t h m 1. To get the distance of vehicles with respect the referenced
vehicle and the angle of existence.
Start
While( angle=A1 to A2 degrees)
while(Vehicle[I]) // vehicle exists
//angle[I][] is the angles of objects or vehicles found in the
//environment obtained from controller.
//distance[I][] is the distances of objects from the reference
//object in the environment measured in meters.
{
Then angle[I][]=A; //A is the angle at which sensor
identifies
// the vehicle that is stepper motor angle
distance[I][]=x;
}
// vehicle(I) is the condition that vehicle exists in
//environment
swap(A1,A2)
goto start
A l g o r i t h m 2. To find the elapsed collision time
Start
Begin
Number of sensors =n
//I is the sensor number
Begin
while(I<n)
{
If (angle[I][]= -2 to 2degrees)
Then case1
Else if (angle[I][]= 178 to 182) case2
Else case3
}
Multiple x the data of different sensors
Find the collision point and time elapse to collision
End
Find the minimum time of collision among the vehicles
Warn the driver with the time span of collision and direction
Goto start
R. V. Krishnaiah, K. Soundara Rajan
104 ISSN 0204–3572. Electronic Modeling. 2008. V. 30. ¹ 2
7. Results. The discrete system is monitored with nine vehicles that is col-
lege buses for a period of 2 months time and is appreciated by the motor vehicle
inspector of local town. The system succeeds to identify the vehicle approaching
with 92% rate and the headlights are on and off. The system fails whenever the
vehicle approaching with dim light that is low intensity or when the head lights
are not functioning properly. Such cases are recognized by the system when they
approach nearer. The micro controller based system is tested with 5 vehicles for
a period of one month (Tables 1—10). This system succeeds with 96% of the
samples taken. This system is not succeeding when in the power supply fluctua-
tions are observed.
The data acquisition and processing of this paper is implemented. The data
of distance and the angle of approaching vehicle is collected from one to five ve-
hicles simultaneously during the period of 15 days with the system installed in
two college buses. The system succeeds with 78% rate. This system advised the
driver regarding hit with voice information of time of collision and angle of col-
lision. The failure is due to the processing speed of data acquisition because of
using only two sensors.
Intelligent Vehicle System in Undefined Environment
ISSN 0204–3572. Ýëåêòðîí. ìîäåëèðîâàíèå. 2008. Ò. 30. ¹ 2 105
X1 (m) X2 (m) t (sec)
73.646114
66.531400
61.327663
63.184064
59.304756
55.288271
42.812571
33.010236
42.532341
116.305632
146.099281
174.625313
164.681511
144.544111
111.928549
86.301796
3
3
2
3
3
3
2
2
Table 2. Output sample for input sample 1
X (m) V1(m/s) Angle of arrival(rad)
250.5
226.3
208.6
185.3
162.8
142.7
110.5
85.2
21.2
20.8
21.1
19.6
18.4
17.5
16.8
15.4
1.45
1.45
1.45
1.43
1.42
1.41
1.41
1.41
Table 1. Input sample 1
X1 (m) X2 (m) t (sec)
151.934283
146.525619
130.159143
109.668412
96.128271
74.437098
57.394034
30.992101
14.081168
108.981650
125.019998
161.728167
173.407036
151.996710
117.698768
90.746901
53.028294
26.305893
7
6
6
5
5
4
3
2
1
Table 4. Output sample for input sample 2
X (m) V1(m/s) Angle of arrival(rad)
240.5
216.3
208.6
185.3
162.8
142.7
110.5
85.2
50.3
21.2
20.8
21.1
19.6
18.4
17.5
16.8
15.4
13.2
1.83
1.82
1.82
1.82
1.81
1.81
1.81
1.81
1.79
Table 3. Input sample 2
The results of the data acquisition and processing are verified with the
results of the mathematical model explained in simulation part. The results are satis-
factory that most of the values are matched with success rate of 72%. Three sam-
ples are shown below. The number of samples of data acquired is 55 related to
different vehicles approaching in different directions. This data involves differ-
ent angles, distances and the velocities.
R. V. Krishnaiah, K. Soundara Rajan
106 ISSN 0204–3572. Electronic Modeling. 2008. V. 30. ¹ 2
X1 (m) X2 (m) t (sec)
239.629020
215.516661
207.844547
184.628929
162.210413
142.183206
110.099820
94.855229
50.117837
25.108737
28.792360
25.903668
24.979992
22.181073
19.493589
17.088007
13.228757
11.357817
6.000000
3.000000
11
10
9
9
8
8
6
6
3
2
Table 6. Output sample for input sample 3
X (m) V1(m/s) Angle of arrival(rad)
240.5
216.3
208.6
185.3
162.8
142.7
110.5
95.2
50.3
25.5
21.2
20.8
21.1
19.6
18.4
17.5
16.8
15.4
13.2
10.2
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
0.12
Table 5. Input sample 3
X1 (m) X2 (m) t (sec)
157.583241
145.257906
136.187629
122.755818
110.298755
85.410038
65.854618
38.639952
19.781931
109.462322
141.534448
171.822583
166.844239
146.342065
113.318136
87.372765
50.418251
25.238859
7
6
6
6
6
5
4
2
1
Table 8. Output sample for input sample 4
X (m) V1(m/s) Angle of arrival(rad)
226.3
208.6
185.3
162.8
142.7
110.5
85.2
50.3
25.2
20.8
21.1
19.6
18.4
17.5
16.8
15.4
13.2
10.2
1.25
1.25
1.23
1.22
1.21
1.21
1.21
1.18
1.17
Table 7. Input sample 4
X1 (m) X2 (m) t (sec)
70.708589
66.575235
62.266207
56.428147
51.337800
30.008085
15.757415
172.971096
166.267856
146.174553
114.594939
99.141313
52.858301
26.608269
3
3
3
3
3
2
1
Table 10. Output sample for input sample 5
X (m) V1(m/s) Angle of arrival(rad)
240.5
216.3
208.6
185.3
162.8
95.2
50.3
21.2
20.8
21.1
19.6
18.4
15.4
13.2
4.69
4.68
4.67
4.57
4.56
4.52
4.50
Table 9. Input sample 5
7. Conclusions. The results of three samples are given. In an environment
the number of samples depends on the number of vehicles. The number of sen-
sors may increase to get better accuracy and decrease the failure rate. The resolu-
tion of sensors may be increased if required for an application. The samples of all
sensors of one rotation are multiplexed to find the minimum of those elapse col-
lision times with different vehicles in the environment. This system may be de-
veloped as rugged system. This will be used as a black box for information gath-
ering if we maintain backup with protection. This facility may be used in other
continents also for legal purpose that which vehicle has violate the rule of traffic,
where the roads are well developed and the collision rate is low.
The future expansion of this paper may be controlling the vehicle based on
the advisory system output. The Control of vehicle involves the mechanical ope-
rations like applying break and steering actions with the help of data supplied
electronically by the controller. This is required particularly for Indian system to
avoid the collisions on the roads because accidents are occurring even though
the speed of vehicles is low. The data of this paper will be helpful for guiding the
intelligent vehicle on the road in undefined environment. This system need not
be installed in other vehicle, which exists in the unknown environment. This is
the advantage of this methodology. Those who require safety measures they may
use it.
Ðîçãëÿíóòî çàäà÷ó ðîçï³çíàâàííÿ ñèòóàö³¿ ç ÷îòèðèêîë³ñíèì òðàíñïîðòíèì çàñîáîì íà äîðîç³ â
óìîâàõ íåâèçíà÷åíîñò³ íàâêîëèøíüîãî ñòàíó, íàïðèêëàä ïðè ïðîõîäæåíí³ òðàô³êà ó ñåëèù³
òà ì³ñò³. Ñïî÷àòêó ñèñòåìó âèêîðèñòîâóâàëè äëÿ çàïîá³ãàííÿ âèõîäó òðàíñïîðòíîãî çàñîáó çà
ìåæè äîðîãè âíî÷³ âíàñë³äîê îáìåæåíîñò³ â³çóàëüíîãî îãëÿäó äëÿ âîä³ÿ. Ñèñòåìó çàïðî-
ïîíîâàíî ÿê äæåðåëî êîíñóëüòàòèâíî¿ ³íôîðìàö³¿ äëÿ âîä³ÿ òðàíñïîðòíîãî çàñîáó.
1. Fay A. A fuzzy Knowledge-based system for railway traffic control// Engineering Applica-
tions of Artificial Intelligence.— 2000.— 13. — P. 719—729.
2. Hilton J., Shankar U. 2001 Motor Vehicle Traffic Crashes Injury and Fatality Estimates
Early Assessment//.DOT HS 809 439. U.S. Department of Transportation, National High-
way Traffic Safety Administration, National Center for Statistics and Analysis. — Washing-
ton: DC, April, 2002.
3. Bellis M. The History of the Automobile. Http://inventors.about.com/library/weekly/
aacarssteama.htm.2002.
4. Henry R. D. Automatic Ultrasonic Headway Control for a Scaled Robotic Car// Thesis. —
Virginia Polytechnic Institute and State University. 2001.
5. Traffic Safety Facts 2000: A Compilation of Motor Vehicle Crash Data from the Fatality
Analysis Reporting System and the General Estimates System// DOT HS 809 337, U.S. De-
partment of Transportation, National Highway Traffic Safety Administration, National Cen-
ter for Statistics and Analysis. — Washington: DC, December, 2001.
6. Reed T. B. Discussing Potential Improvements in Road Safety: A Comparison of Conditions
in Japan and the United States to Guide Implementations of Intelligent Road Transportation
Systems// IVHS Issues and Technology. — 1992. — SP- 928. — P. 1—12.
Intelligent Vehicle System in Undefined Environment
ISSN 0204–3572. Ýëåêòðîí. ìîäåëèðîâàíèå. 2008. Ò. 30. ¹ 2 107
7. Utter D. Passenger Vehicle Driver Cell Phone Use Results from the Fall 2000 National Occu-
pant Protection Use Survey. Research Note//DOT HS 809 293. U.S. Department of Trans-
portation, National Highway Traffic Safety Administration. — Washington: DC, July, 2001.
8. Krishnaiah R. V., Dr. Rajan E. G. A fast algorithm for collision avoidance by autonomous
mobile systems in the frame work of cellular logic array processing// National Conference on
Intelligent Manufacturing Systems. — Coimbatore Institute of Technology. — Coimbatore.
February, 1997. — P. 02.1-1 — O2.1-5 .
9. Harlow.C, Peng.S. Automatic vehical classification with range sensors// Transportation Re-
search. — 2001. — Part C9. — P. 231—247.
10. Goerick C., Noll D., Werner M. Artificial neural networks in real-time car detection and
tracking applications// Pattern Recognition Letters. — 1996. — 17. — P. 335 — 343.
11. Wu W., Qisen Z., Mingjun W. A method of vehicle classification using models and neural
networks// Proceedings of the 53rd IEEE Vehicular Technology Conference. — Rhodes.
Greece, May 6—9, 2001. — P. 3002—3026.
12. Nooralahiyan A. Y., Kirby H. R., Mckeown D. Vehicle classification by acoustic signature//
Math.Comput.Modelling. — 1998. — 27, ¹ 9 — 11. — Ð. 205—214.
13. Pasquariello G., Satalino G., La ForgiaV., Spilotros F. Automatic target recognition for na-
val traffic control using neural networks// Image and vision computing. — 1998. — 16. —
Ð. 67—73.
14. Dr. Pushkin Kachroo, Chair, Dr. A Lynn Abbott, Dr. Hugh VanLandingham Feedback Con-
trol for a Path Following Robotic Car //Thesis note. Virginia Polytechnic Institute and State
University. April, 2002. — P. 2—3.
15. Fritz H. Model-based neural distance control for autonomous road vehicles// Proceedings of
the 1996 IEEE Intelligent Vehicle Symposium. — Tokyo, Japan, September 19—20. —
1996. — Ð. 29—34.
16. Lauffenburger J. Ph., Basset M., Coffin F., Gissinger G.L. Driver-aid system using path-
planning for lateral vehicle control // Control Engineering Practice. — 2003. — 11. ¹ 2. —
Ð. 217—231.
17. Touran A., Brackstone M., McDonald M. A collision model for safety evaluation of autono-
mous intelligent cruise control // Accident Analysis and Prevention. — 1999. — 31. —
Ð. 567—578.
18. Von Seelen W., Curio C., Gayko J., Handman U., Kalinke T. Scene analysis and organization
of behavior in driver assistance system// International Conference on Image Processing. —
Vancouver, Canada, September 10—13. — 2000. — Ð. 524—527.
19. Wahle J., Annen O., Schuster Ch., Neubert L., Schreckenberg M. A dynamic route guidance
system based on real traffic data// European Journal of Operational Research. — 2001. —
131. — Ð. 302—308.
Ïîñòóïèëà 14.03.07
R. V. Krishnaiah, K. Soundara Rajan
108 ISSN 0204–3572. Electronic Modeling. 2008. V. 30. ¹ 2
|