Design of GPS-Based System to Avoid Camel-Vehicle Collisions: A Review
Mohammed S. Zahrani,
Asrar Ul Haque
Hundreds of Camel-Vehicle accidents are reported every year causing numerous deaths and loss of property running into millions of Saudi Riyals. Ministry of Transport of the Kingdom spends billions of Saudi Riyals to deal with this problem by building fences along the highways which costly, difficult to maintain and isolate the camel habitat. To address this problem, a deployable and intelligent Camel-Vehicle Accident Avoidance System (CVAAS) was designed using global positioning system (GPS) technology. This paper conducted a survey of the developed animal detection and warning systems. It provided recent and numerous reviews of the worldwide technologies which are being used to reduce animal-vehicle collisions. Also, this paper introduced the design of the Camel-Vehicle Accident Avoidance System (CVAAS) in the Kingdom of Saudi Arabia. The CVAAS project took a first comprehensive step towards a system that will help detect camels on the highway and warn drivers as well. The innovative findings of this study is the careful and intelligent use of the GPS to detect the camel position, direction and movement. Moreover, CVAAS identified the dangerous zones that enables the warning system to adapt the alarming period. The CVAAS can be classified as an Intelligent Transportation System (ITS). The use of GPS technology in this kind of application is a novel idea. This proposal can be recognized as a world leader in using GPS for avoiding animal-vehicle collision. The research highlighted the advantages of CVAAS to save human lives as well as camels in such accidents and save billions of Saudi Riyal.
August 13, 2010; Accepted: November 23, 2010;
Published: February 22, 2011
Hundreds of Camel-Vehicle accidents are reported every year causing numerous
deaths and loss of property running into billions of Saudi Riyals. Al-Ghamdi
and Al-Gadhi (2004) summarized traffic accident data and reported that more
than 600 camel-vehicle accidents occur annually. Usually camels found near the
highway are domestic camels because the owners like to live close to highway
for transportation facility. These camels move across highways looking for water
and food and during mating season. Camels are very hard to detect by vehicle
drivers especially during the night and results in severe accidents. Presently,
in Saudi Arabia, there is no warning system installed for the vehicle drivers
to slow down when there is a camel on or near the highway.
Currently fences are used to stop camels from coming on the highways. Fences
are expensive to install and difficult to maintain. Also, these fences divide
the camel habitat and isolate camel populations. Furthermore, it is known that
camel owners create holes in the fences to allow camels to across them. This
kind of behavior literally nullifies the effect of fences and leads to accidents.
Therefore, a warning system is needed to inform the vehicle drivers when a camel
is on or near the highway so that drivers can respond by slowing down and avoiding
accidents (Al-Ghamdi and Al-Gadhi, 2004).
Many kinds of animal detection and warning systems are used around the world
to indicate presence of animals on highways to avoid accidents. Animal detection
systems are divided into three main categories namely Road-based, Vehicle-based
and Animal-based. The use of such animal detection and warning systems has shown
to reduce the number of accidents. However, such systems have various drawbacks
including generating false detections. The main objective of this research was
to design, simulate and deploy a Camel-Vehicle Accident Avoidance System (CVAAS)
using global positioning system (GPS) technology. The GPS is a free service
which provides exact geo-location (accuracy <3 m) of a GPS receiver by the
use of a network of 24 satellites. The use of GPS receivers has increased tremendously
for navigation purpose and in tracking animals (Macklean,
2009; Rutter, 2007; Turner et
al., 2001), in sensors networks and many other applications. The GPS
receiver can be obtained for a reasonable price of around 80-200 SAR. GPS devices
can be integrated with Geographic Information System (GIS) software packages
for further programming and developing applications such as CVAAS. The use of
GPS is a novel as well as feasible solution to develop an animal detection and
The CVAAS system is an animal-based system which primarily makes use of a programmable GPS receiver and a warning system. The system will identify the presence of a camel on or near the highway and then the programmable GPS device will send out a signal to the Dedicated Short-Range Communication (DSRC) transmitter. Consequently, the DSRC transmitter forwards the camel position to a DSRC receiver mounted on a warning system. The signal will activate the warning system to warn the vehicle drivers to slow down in order to avoid collision with the camel. Figure 1 illustrates such a scenario.
|| The Warning System being Activated as Camels approach the
In Saudi Arabia, motor vehicles are the most common mode of transportation with comparatively good roads passing through extensive deserts.
Arabian camels having one hump, also known as dromedaries, is the traditional
animal of the desert. They can be over 2.1 m tall at the hump and weigh up to
726 kg (National Geographic, 2009). The humps storing
up to 36 kg of fat give camels their ability to travel as far as 161 desert
kilometers without water. Camels are capable of losing 40% of its body's weight
before becoming distressed, they are able to go five to seven days before having
to drink (Harris, 2005). Al-Amr et
al. (1998) found that camel have very strong sense of smell which they
use to locate water from a distance of 2.5 km and recognize another camel from
11 km. They can move 50 to 60 km looking for food. They can travel as long as
20 days to return to their places. A female camel has a strong emotion to go
back to her birth place along with her calves. The pregnancy in female camels
lasts 390 days in the one-humped camel (FAO, Animal Health
Manual, 1994). They can run at up to 65 km h-1 in short bursts
and sustain speeds of up to 40 km h-1. Studies conducted in Australia
showed that camels have tendency to race with a car when they see one (Al-Amr
et al., 1998). The camel is the traditional animal of the desert
and the camel population increases continuously. For example the number of camels
in UAE increased from 100,000 to 250,000 in the last 40 years (David,
2007). More than half a million camels move freely in Saudi Arabia (Al-Ghamdi
and Al-Gadhi 2004). Around Riyadh and Qassim area, the density of camels
concentration per square km was reported 0.4 and 0.6 camels (FAO,
2004). Al-Hazmi and Al-Bar (1999) reported that
the density of camels along the highways in western areas in Saudi Arabia is
12 Camels/Kilometer. Because of their instinct behavior, camels travel long
distances and cross a number of rural highways for food and water. Therefore,
collisions with camels are more prominent (Al-Sebai and
Al-Zahrani, 1997; Ansari and Ashraf, 1998). The
rate of accidents is higher during nights as camels that stay in the desert,
mostly unsupervised, move around in herds, often coming on to the roads without
warning. There are no designated crossing points for the camels and the fences
erected in various places is often mutilated by camel owners to enable the camels
to cross over (Ansari and Ashraf, 1998). Camels usually
move as a herd on the road and this will give no space or chance for the driver
to avoid the collision. They reviewed 140 patients with low cervical spinal
cord injuries in the Riyadh Armed Forces Hospital. Motor vehicle accidents constituted
119 (85%) of the patients while the camel collisions were a major cause of vehicle
accidents 39 (33%), as reported by Ansari and Ashraf (1998).
According to the study of Al-Ghamdi and Al-Gadhi (2004),
the most frequently involved animal in AVCs is camel; it is estimated
that 97% of all reported AVCs were camel related. More than 90% of these
accidents occur at night, between dusk and dawn (Al-Amr et
al., 1998). These accidents cause a lot of damage to the environment,
economy and social life such as significant economic loss, human injuries and/or
fatalities, loss of valuable wildlife and damage to properties (Fig.
2). Al-Ghamdi and Al-Gadhi (2004) studied the incidents
of camel-crossing related accidents on Saudi rural roads to develop techniques
to deal with the problem. In the study, seven camel-crossing warning signs were
used to find out if they were effective in reducing the number of camel-vehicle
collisions. The measure of effectiveness utilized the reduction in speed of
the motorists passing by the signs. Although the signs brought about reductions
in speed, they were relatively ineffective with drivers slowing down by only
three to seven kilometers per hour, the study showed.
The AVC is not only a traffic problem in Saudi Arabia but also considered a
major safety problem in USA, Japan and Europe (Bruinderink
and Hazebroek, 1996; Stout et al., 1993).
In Europe, excluding Russia, an estimated 507,000 ungulate-vehicle accidents
occur annually (Bruinderink and Hazebroek, 1996).
|| Camel-Vehicle accident
The number of AVCs continually increased during the last 30 years with
the increase in traffic volumes, vehicle speed and animal populations. In 1980
approximately 200,000 deer were killed on U.S. highways in deer-vehicle collisions
(DVCs) and by 1991 the total number of deer killed as a result of deer-vehicle
collisions were 538,000 based on collision data from only 36 states. Currently,
It is estimated that approximately 726,000 deer-vehicle accidents occur annually
in the U.S. (Danielson et al., 1998). In 1998,
an estimate of approximately 13,500 DVCs occurred in Iowa alone (23% increase
in 3 years) resulting in an annual total damage of more than $ 10 million in
personal injury and property damage to Iowa vehicles.
In the United States, considerable interest has been generated preventing AVCs
due to the fact that over a billion dollars worth of vehicle damage annually
(Hedlund et al., 2004). It has been noted that
the large proportion of hospitalization crashes (44.8%) found involving drivers
trying to avoid an animal, resulting in the vehicle leaving the road, hitting
a tree, pole, fence, or rolling over (Conn et al.,
2004). Langley et al. (2006) examined the
risk factors involved with fatal AVCs in the United States from 1995 to
2004 and found that 89.5% occurred on rural roads, 64.8% in darkness, 85.4%
on straight sections of road, 91.1% occurred in dry weather conditions and 28%
of the victims were motorcyclists. A large proportion of injuries from AVCs
in the United States involve deer (Sullivan et al.,
2004). Similarly, in Europe and Canada moose and deer have been shown to
be a considerable problem on the road (Haikonen and Summala,
However, many countermeasures to prevent animals from entering the roadway
or to change driver behavior were trialed but the majority were found to have
generally very limited effect, particularly for deer in the United States (Hedlund
et al., 2004).
These countermeasures include driver education, warning signs, roadside reflectors,
deer repellents, lower speed limits and deer whistles. However, some studies
found positive results. For example, Sullivan et al.
(2004) found that temporary warning signs that were well lit at night during
deer migration periods reduced the number of deer road-kills at experimental
sites compared to control sites. However, the study did not report human casualties
during the trial.
Review of technologies: Worldwide numerous technologies have been used in attempts to reduce big AVC. This section classifies the techniques used in the past to reduce AVC into three categories roadway-based, animal-based and vehicle-based techniques. The first category (roadway-based technique) was developed for and dedicated to the highways. It includes roadway fencing, underpasses, overpasses, roadway reflectors, warning signs, infra-red, microwave RF, etc. Roadway-based detection systems, however, are designed to inform all drivers, regardless of what equipment their vehicle may or may not have. The second category (animal-based) includes those technologies which were installed in animals to reduce the AVC. The third category (vehicle-based) includes the technologies that are equipped into vehicles to reduce the AVC.
Roadway-Based conventional techniques
Fences: Fences have been installed to keep animals away from the road (De
Molenaar and Henkens, 1998; Clevenger et al.,
2001). Roadway fencing is the famous conventional techniques used to reduce
AVC. Ward (1982) signified that a 2 m high big-game
fence is effective in reducing vehicle collisions involving deer. Fencing is
extremely expensive because they have been combined with wildlife crossing structures
such as underpasses and overpasses that enable animals to move freely along
both sides of the highways. Fencing must be inspected frequently and repaired
to original condition to be successful at reducing collisions because animals
quickly exploit breaks in the fence (Foster and Humphrey,
1995). Apparently, deer continually test fencing, making a good maintenance
program necessary (Ward, 1982). Reed
et al. (1982) estimated that maintenance cost for fencing was approximately
1% of construction cost per year. Moreover, fencing without crossing structure
isolates animals' populations.
Warning signs: Signs that warn drivers of high big animals (e.g. deer)
-crossing probabilities are the most common approach to reducing AVC (Putman,
1997). Romin and Bissonette (1996) suggested that
deer crossing signs may be effective if drivers would reduce their vehicle
speed. Lighted, animated deer-crossing warning signs were evaluated in Colorado.
Pojar et al. (1975) concluded that drivers
speeds were initially slightly reduced, but after the drivers got used to the
animated sign, it lost its effect in reducing the drivers speeds.
Highway lighting: Most of the AVC occurred from sunset to sunrise. It
was expected that highway lighting enhance drivers' night vision and reduce
AVC. Highway lighting did not affect drivers behavior or animal crossings-per-accident
ratios (Reed, 1981). Thus, increased highway lighting
was not effective at reducing AVC.
Roadway-based detection systems: Animal detect technologies detect large
animals as they approach the road. When an animal is detected, signs are activated
that warn drivers that large animals may be on or near the road at that time.
Vehicle detect technologies operate on a slightly different principle as they
detect vehicles, not the animals. They detect vehicles or trains, not the animals.
Once a vehicle or train is detected large animals are alerted through a range
of audio and visual signals from stations placed in the right-of-way (Huijser
and McGrowen, 2003). Briefly, different technologies used to develop animal-detect
and vehicle-detect techniques are summarized below.
First, a series of passive infrared sensors were designed and installed in
seven sites in Switzerland to detect deer within 30-100 m radius on both sides
of the road. Once a deer was detected LED signs with a deer symbol were activated
and stayed on for 45 seconds to alert the drivers (Kistler,
1998; Clevenger et al., 2001). This technique
produced false detections because of strong winds and warm engines of passing
vehicles. Roadway fencing is one of the famous conventional techniques used
to reduce AVC. Ward (1982) signified that a 2m high
big-game fence is effective in reducing vehicle collisions involving deer. Fencing
is extremely expensive because they have been combined with wildlife crossing
structures such as underpasses and overpasses that enable animals to move freely
along both sides of the highways. Fencing must be inspected frequently and repaired
to original conditions to be successful at reducing collisions because animals
quickly exploit breaks in the fence (Foster and Humphrey,
1995). Apparently, deer continually test fencing, making a good maintenance
program necessary (Ward, 1982). Moreover, the broken
sensors, loss of power due to snow covered solar panels and broken lamps in
the warning signs caused additional problems. Similarly, the Flashing Light
Animal Sensing Host (FLASH) was designed to detect mule deer in Wyoming, USA.
It also used a series of infrared sensors (Gordon, 2001).
More than 50% of the detections through FLASH system were false. This was due
to frost on the sensors, birds feeding on carrion in the crossing area and snow
thrown by passing snowplows.
Second, in Finland, microwave radar sensors that were designed and installed
to detect large animal (e.g. moose) movements up to 50 m in distance within
a 60° horizontal angle. When a large animal was detected, LED message signs
with a moose symbol were turned on and remained on for two-three minutes (Pynn
and Pynn, 2004). To verify the presence of large animals a video camera
is installed. In addition, to distinguish moose from other moving objects such
as rain or rain spray, the system was programmed to only detect objects moving
towards the sensors at a speed greater than 0.8 m sec-1. This technique
produced false detection in spring when the snow melted and the water warmed
on the pavement, spray from passing vehicles triggered the system.
Third, in 2000 an animal detection system was installed in Washington, USA.
It consisted of two lasers, one placed on each side of the road, two standard
deer warning signs, two smaller rectangular signs that read When Flashing and
two solar-powered red flashing beacons. When the laser beam was broken the lights
were switched on. The lasers operated on batteries with a one-week lifespan
while the red strobes were solar powered. The sighting of the lasers proved
difficult, partly because of the distance between the sensors. Sunlight heating
up of the plastic boxes holding the laser equipment may have caused problems
with the sighting of the laser (Shipley, 2001). False
detections caused the batteries to drain quicker than anticipated. Finally,
the system was taken down spring (Shipley, 2001). Similarly,
in October 2002, an animal detection system was installed along US 97A, near
Wenatchee, Washington (Al-Amr et al., 1998). It
used laser beams to detect animals (e.g., deer). If these deer stay there longer
than one minute, the warning signals were turned off and drivers are no longer
warned of their presence (Romin and Bissonette, 1996).
Fourth, an animal detection system based on microwave technology was installed
in 2002 along the highway in Montana (Evink et al.,
2002). It consisted of series of transmitters and receivers. Each transmitter
sent a uniquely coded, continuous microwave RF signal to its intended receiver
(STS, 2002). The transmitters and receivers were mounted
about 120 cm above the ground (designed to detect elk). If this signal was blocked,
the receiver sent a UHF radio signal to the master station. The master station
then sent the beacon-on command to the three nearest beacons. Each beacon was
situated above a standard elk warning sign and signs that showed, when flashing
next 1 mile. The flashing beacons alerted on-coming traffic that there might
be a large animal on or near the road. After the designated timeout period (3
min), the master station transmitted the beacon-off command to the beacon stations.
If the signal was blocked continuously, the beacons would stop flashing after
12 min. The system produced a large number of false detections for several causes
such as snow spray.
A vehicle detect system was in April, 2002 in Canada. It consists of a small
cabinet with electronics, sensors for vehicle detection and an animal warning
device. When no vehicles are present, the system is not active. Once vehicles
are detected, units in the roadside are activated that alert deer through a
variety of noise and light signals (IRD, 2002). Therefore,
the audio and visual signals produced by the stations in the right-of-way may
not scare the animals away from the road once they have been exposed to it for
a certain time. Additionally, such system is not well suited for high traffic
flows since the animal warnings would be running continuously in such locations.
Animal-based technologies: The animal based technologies, to avoid animal collision, used different types of collars fasten with the animal to trigger a warning system such as blinking signals. They are classified as reflective collars and radio collars.
Reflective collars: In British Columbia, Canada, the ministry of environment
conducted a method to reduce AVC. They put collars with reflective tape on a
number of animals to increase their visibility to drivers (Tan
and Huang, 2006).
Radio collars: Multiple projects utilized radio collars since 1999 up
to now. A system was installed along a 4,827-m-long section of Hwy 101, near
Sequim, on the Olympic Peninsula,Washington. In 1999 about 10% of the elk herd
was radio collared (Carey, 2001). An effort was made
to radio collar lead cows, but this was not always possible. Receivers placed
along the road scan for the frequencies of the individual radio collars 24 h
per day. When the radio-collared individuals come within about 400 m of the
road, the receivers that pick up the signal activate the flashing beacons that
are linked to that receiver. There are four receivers in total. Typically only
one receiver picks up the signal at the same time, but if the radio-collared
individual is about halfway between two receivers, the signal may be picked
up by both receivers. Two receivers are linked to only one flashing beacon (at
both ends of the road section). The two other receivers are each linked to two
flashing beacons, one for each travel direction. To block false detections,
a device that counted the pulses of the radio signal had to be added. This device
filtered out signals from other, non-elk, radio transmitters. The system became
operational in fall 2000. The batteries of the radios have a three-year life
span, but most of them last much longer. A second capture session took place
in March 2003. There were eight elk (7 cows, 1 bull) with a radio collar in
September 2003. The system seems to work well, even after a change in habitat
use caused the elk to cross the road more frequently than they used too.
Radio-collar systems, such as the one in Sequim, can also produce false negatives.
It is unlikely that all the individuals in a certain area can be equipped with
radio collars. As a consequence, the animals without a radio collar are only
detected if radio-collared animals accompany them. Therefore, the system only
works well for highly gregarious species. The system also works much better
for a resident population than for migrants from far-away locations that may
only cross the road once or twice per year. The radio-collar system requires
re-collaring effort. In addition, individuals may die as a result of hunting,
injuries or old age. Experts usually minimize the stress for the animals during
capturing and handling, but the animals are exposed to a certain amount of stress
during capturing and handling and as a result of carrying a radio collar.
Wildlife collisions are a worldwide problem and WSDOT is one of hundreds of
agencies that have been trying for years to find effective and affordable ways
to reduce them. WSDOT put radio collars on elk near Port Townsend that trigger
flashing beacons telling drivers there are animals on the road ahead. It was
found that when a collared elk settles down for the night near a receiver, the
beacon flashes all night. Radio collars and capturing the elk to put them on,
is also very expensive (Carey, 2001). This project deals
with camels, domestic animals that needless of hunting efforts and the maintenance
can be done easily by camel owners. A lesson was attained from the WSDOT that
it is required to find a way to avoid continuous alarming when a collared camel
settles down for the night near a receiver.
On the other hand, GPS collar is a valuable tool for documenting the movements
of large, wide-ranging animal kinds. It provides abundant amounts of continuous
movement data amid even extreme environmental conditions, while greatly reducing
the number of man-hours required in the field. Recently, GPS collar has been
instrumental in monitoring large mammals use of highways and wildlife underpasses
in Arizona (Mckinney and Smith, 2007; Dodd
et al., 2007; Gagnon et al., 2007).
Using data gathered from GPS collar, Mckinney and Smith (2007)
were able to identify spatial patterns in bighorn sheep movement relative to
a key section of US 93. Based on GPS collar data, the authors were able to make
recommendations regarding placement of wildlife-engineered crossing structures
on US 93. Dodd et al. (2007) used GPS collars
to assess permeability of SR 260 to elk through successive phases of reconstruction,
which included widening the highway, integrating wildlife crossing structures
and implementing ungulate-proof fencing. Gagnon et al.
(2007, 2009) were able to determine how patterns
in traffic flow affected elk crossing and distribution in the vicinity of SR
260; the authors found that although high traffic volumes greatly affected elk
crossings, seasonality and proximity to quality habitat also strongly affected
Vehicle-based technologies: Vehicle-based technologies (typically infrared
detectors) inform drivers when a large animal is detected within a certain range
from the sensors attached to the vehicle (Bendix Commercial
Vehicle System, 2004; Hirota et al., 2004;
Honda Motor Co. Ltd., 2004). The range should be sufficient
to allow for the driver to stop the vehicle before impacting the detected animal.
The system could potentially detect large animals anywhere; it would not depend
on the installation of any roadside equipment. However, it is uncertain whether
these on-board detectors are still in production.
Communication based technologies: Numerous studies have been conducted
in the field of communication using different tools for information transfer
from one individual to another i.e., person-person or person to animal. Ragab
et al. (2004a) reported that to meet the increasing demand for real-time
content delivery, the proposed Autonomous Community Information System offers
an efficient information dissemination infrastructure with a decentralized architecture.
The aim of ACIS was to help end-user communicate and exchange information efficiently.
In order to meet this goal, the system uses an application-level multicast technique
that arbitrarily scales to large groups. The ACIS system also features a scalable
community-construction and maintenance scheme that eases the burden of organizing
an online community network.
Ragab et al. (2003a) and Ragab
et al. (2004b) reported that Autonomous Community Information System
(ACIS) is a proposition made to contend with the extreme dynamism in the large-scale
information system. ACIS is a decentralized bilateral-hierarchy architecture
that forms a community of individual end-users (community members) having the
same interests and demands in somewhere, at specified time. It allows the community
members to mutually cooperate and share information without loading up any single
node excessively. In this paper, an autonomous decentralized community communication
technique is proposed to assure a flexible, scalable and a multilateral communication
among the community members. The main ideas behind this communication technique
are: content-code communication (community service-based) for flexibility and
multilateral benefits communication for scalable and productive cooperation
among members. All members communicate productively for the satisfaction of
all the community members. The scalability of the system's response time regardless
of the number of the community members has been shown by simulation. Thus, the
autonomous decentralized community communication technique reveals great results
of the response time with continuous increasing in the total number of members.
In an other study, Ragab et al. (2003b) stated
that the originality of the proposed communication technology dos not come only
from the content-based communication but also from the reply-all that satisfies
the multilateral benefits. In 1 to N community communication all members cooperate
for the satisfaction of all community members contrary to the peer-peer (P2P)
communication techniques. They concluded that the community is: service-based,
cooperative, relationship and multilateral benefits communication.
Ragab et al. (2002) concluded that conventional
communication technologies use the destination address (e.g. unicast address,
multicast address) to send the data. In very changing environment likes ACIS
(i.e. end-users are frequently joined and left), these conventional communication
technologies are not applicable. Thus, the autonomous decentralized community
communications technology has broached to assure a productive cooperation and
a flexible and timely communication among members. In this communication technology,
the sender does not specify the destination but only sends the content/request
with its interest code (CC) to its neighbor nodes.
RESULTS AND DISCUSSION
The previous sections shows that diverse types of systems have been installed
around the world to reduce AVC. Many of the systems encountered technical problems
and maintenance issues. More importantly they experienced false positives and
false negatives. The false positive occurs when the warning system is activated
even if there is no animal. Whereas, false negative occurs when there is animal
but the warning system is not activated. It is noteworthy that previous animal
based detection systems, to which CVAAS fits in, have shown to produce less
false positives and false negatives. False positives may cause drivers to eventually
ignore activated signs (Gordon and Anderson, 2002) and
false negatives present drivers with a hazardous situation. Drivers responses,
through reducing vehicle speed or increased alertness, determines how effective
animal detection systems really are. It is of immense importance that any system
designed to reduce or avoid AVC should ensure minimal number of false positives
and false negative. The CVAAS aims to address these false detection problems
by using the novel idea of using GIS along with GPS which gives accurate positioning
of an animal. Previously the GIS incorporated with GPS technologies were used
to monitor and collect data for the migration and movements of the animals (Rutter,
2007; Turner et al., 2001). An integrated
system of GIS and GPS is a novel idea in designing a system for avoiding domestic
animal and vehicle collisions. In the following section we discuss the high
level design of CVAAS.
Design of Camel-Vehicle-Accident-Avoiding-System (CVAAS): The design
of CVAAS consists of two sub-systems: Animal Detection sub-Systems (ADS) and
Warning sub-System (WS) as shown in Fig. 3. ADS includes two
units: Animal-Based Unit (ABU) and Road-Side Unit (RSU). ABU is attached to
the animal and consists of GPS receiver, DSRC transmitter and interface as shown
in Fig. 4. The European Telecommunications Standard Institute
(ETSI) decided to allocate frequency band from 5875 to 5905 MHz for ITS (ETSI,
2008). Similarly, it was decided to utilize DSRC transmitter and receivers
that operate with 10MHz band 5.885 to 5.895 GHz in CVAAS.
They operate with 33 dBm@10 MHz transmit power that enable transmitters and receivers to reach communication distance range from 500 to 1000 m.
|| Block diagram of Camel-Vehicle-Accident-Avoiding-System (CVAAS)
|| Components of Camel-Vehicle-Accident-Avoiding-System (CVAAS)
|| Description of dangerous zones
The GPS receiver operates to capture key data such as animals position, velocity, acceleration, heading, etc. The transmitter forwards that key data to the RSU. The ABUs interface grants the ability to update the system parameters of both GPS receivers and DSRC transmitter such as the frequency of key data transmission, positioning times based on animal behavior (e.g. more frequent during activity, less frequent when relaxing), packet payload size and message life time. Road-Side units are organized along the highway. Each RSU consists of DSRC transmitter and receiver, communication unit, processing unit (Application processor) and storage area. It stores the highway map for a distance of 100-300m around the RSU and the description of the dangerous zones around (Fig. 5).
The RSUs receiver gets the key data from ABUs transmitter. The communication unit forwards the received key data to the RSUs processing unit. The processing unit executes a thread that runs the RSU-activate procedure. As soon as, the RSU-activate procedure receives the key data that matches with the description of the dangerous zones it takes the decision to send activate message to the WS. The activate message includes the classification of dangerous zones. The WS executes a thread that runs the Warning setup procedure. When the Warning setup procedure receives the activate message, it identifies the degree of hazardous and setups the alarming period. For example, red-zone is the most dangerous zone that includes bi-direction lanes and stripes around it with range10-20 m. If the key data received from the ABU matched with the definition of the red-zone then the RSU activates the WS to produce continuous alarm until receives different data key from the ABU. Otherwise, the WS will not be activated whereas the key data matched with the definition of the green-zone.
Various alternatives to implement warning systems are summarized below.
Warning systems: Different studies were conducted to determine the effectiveness
of intelligent alerting systems for warning drivers of impending collisions.
These include examining the effectiveness of such warning systems on different
age groups (Maltz et al., 2004), as well as on
comparing the effectiveness of alarm warnings presented through different modalities:
visual (GMS, 2002), sound (Tijerina
et al., 2000) and multi-staged alerts (Lee and
Hoffman, 2004). Further research has also been conducted on drivers
performance in distinguishing between sound alarms like auditory icons and beeps
(Graham, 1999), as well as the effectiveness of visual
warnings like warnings signs (GMS, 2002). These studies
comprehensively studied a number of critical issues in the introduction of intelligent
predictive alarms into the driving domain. The presentation of different alerts
could affect drivers performance, as well as the interactions among the
various other alerts and distractions combine to affect situational awareness
of drivers. Recently, an alternative approach known as cooperative driving appeared
based on vehicle-to-infrastructure (V2I) or vehicle-to-vehicle (V2V) communication
(Tan and Huang, 2006; Kais et
al., 2005; Luo and Hubaux, 2004). More specifically,
this new approach of collaborative driving lies in the fact that the infrastructure
or the vehicle can communicate its information (e.g. location) to surrounding
vehicles or nearby infrastructure. In this case, the warning system is intended
by incorporating the information communicated from the surrounding vehicles
into the warning process.
Accordingly, one of the main objectives of the proposed system CVAAS was to
provide a method and a system for warning drivers of the presence of camels
and other large animals near to highways. The present invention provides a warning
system to drivers to notify them of the presence of camels or other large animals
along the highway, such that the drivers can slow down and drive cautiously
to avoid an accident with the camels. The system includes a plurality of Road-Side-Unit
(RSU) posts installed along the highway, with each post having a DSRC receiver
for receiving signals that indicate the presence of camels. Each RSU is operatively
connected to a warning system which is activated when a camel enters the danger
zone. The Warning System (WS) and RSU can be connected to a battery which is
coupled to a solar panel or to a power supply for recharging. The WS can be
electrically connected in series such that all the lights are illuminated when
one of the RSUs senses a camel. The warning system may also include some devices
that are functional only at night (i.e. a light-sensitive photoelectric eye).
The proposed warning system (CVAAS) may be innovative with significant technological
contributions which involved:
An automatic detection and relevance check of traffic through a global-positioning-system (GPS);
||An organized vehicle-to-vehicle communication system for establishing
a communication network with both the on-coming and following vehicles.
Thus, a self-organized overlay network of vehicles is required to be implemented;
||In-vehicle message management and warning dissemination;
||A warning message management for transmission and distribution
of hazard warnings, ensuring driver information in time at the right spot.
||Holographic images can be used which contain visual information
(warnings or signs) for installation as optical barriers in highways. As
the drivers approach the danger area the holographic signs would inform
them, visually, to stop (Orazem, 1995;
Blanche et al., 2010).
This study conducted a survey of the developed animal detection and warning systems. It provided recent and numerous reviews of the worldwide technologies which are being used to reduce animal-vehicle collisions. Also, this paper introduced the design of the Camel-Vehicle Accident Avoidance System (CVAAS) in the Kingdom of Saudi Arabia. The CVAAS project took a first comprehensive step towards a system that will help detect camels on the highway and warn drivers as well. The innovative findings of this project is the careful and intelligent use of the GPS to detect the camel position, direction and movement. Moreover, CVAAS classified the dangerous zones that enables the warning system to adapt the alarming period.
The authors would like to thank KACST (King Abdulaziz City for Science and Technology) for sponsoring this project. The authors also acknowledge the academic support from the College of Computer Science and Information Technology, King Faisal University.
Al-Amr, S.A., S. Al-Hathlool, A. AlGadhi, A.S. Al-Ghamdi and K. Al-Kahtani, 1998. Studying the impact of camels on traffic safety. King Abdulaziz City of Sciences and Technology, Research Project No. AT-14-72, Final Report. Riyadh, Saudi Arabia.
Al-Ghamdi, A.S. and S.A. Al-Gadhi, 2004. Warning signs as countermeasures to camel-vehicle collisions in Saudi Arabia. Accid. Anal. Prev., 36: 749-760.
Al-Hazmi, M. and H. Al-Bar, 1999. Mitigation of unguarded wild animal road accidents through consideration of their biological and behavioral nature. J. King Abdulaziz Univ., 11: 3-21.
Direct Link |
Al-Sebai, M.W. and S. Al-Zahrani, 1997. Cervical spinal injuries caused by collisions of cars with camels. Injury, 28: 191-194.
Ansari, S. and A.K.S. Ashraf, 1998. Camel collision as a major cause of low cervical spinal cord injury. Spinal Cord., 36: 415-417.
Bendix Commercial Vehicle Systems, 2004. Bendix uses infrared vision (Safety systems). Bendix Commercial Vehicle Systems Launches Xvision Infrared Camera. Advanced Transportation Technologies News. February 01, 2002.
Blanche, P.A., A. Bablumian, R. Voorakaranam, C. Christenson and W. Lin et al., 2010. Holographic three-dimensional telepresence using large-area photorefractive polymer. Nature, 468: 80-83.
Bruinderink, G.W.T.A. and E. Hazebroek, 1996. Ungulate traffic collisions in Europe. Conserv. Biol., 10: 1059-1067.
Direct Link |
Carey, M., 2001. Addressing wildlife mortality on highways in Washington. Proceedings of the International Conference on Ecology and Transportation, September 24-28, 2001, Keystone, Colorado, pp: 605-610.
Clevenger, A.P., B. Chruszcz and K. Gunson, 2001. Highway mitigation fencing reduces wildlife-vehicle collisions. Wildlife Soc. Bull., 29: 646-653.
Direct Link |
Conn, J.M., J.L. Annest and A. Dellinger, 2004. Nonfatal motor-vehicle animal crash-related injuries in United States from 2001-2002. J. Safety Res., 35: 571-574.
Danielson, B., M. Hubbard, D. Murray and D. van Helden, 1998. A proposal to develop a deer-vehicle collision reduction initiative. Department of Animal Ecology, Iowa State University, pp: 10.
David, G., 2007. Overgrazing their welcome. Zawaya, 11: 30-33.
De Molenaar, J.G. and R.J.H.G. Henkens, 1998. Effectiviteit van wildspiegels: Een literatuurevaluatie. IBN Rapport 362. Institute Voor Bos-en Natuuronderzoek, Wageningen, The Netherlands.
Dodd, N.L., J.W. Gagnon, S. Boe, A. Manzo and R.E. Schweinsburg, 2007. Evaluations of measures to minimize wildlife-vehicle collisions and maintain permeability across highways: Arizona route 260. Final Report to Arizona Department of Transportation, ECS File No. JPA 01-152 JPA 04-024T.
ETSI, 2008. European telecommunications standard institute, TC Satellite Earth Stations and systems (TC-SES). Broadband Satellite Multimedia (BSM) Working Group Meeting No. 37, ITSI, Sophia Antipolis, France, 16-18 September 2008.
Evink, G., K.P. McDermott and N.C. Raleigh, 2002. The center for transportation and the environment. Annual Report 2002, North Carolina State University, USA.
FAO, 2004. The global livestock production and health atlas (GLPHA). http://kids.fao.org/glipha/.
FAO, Animal Health Manual, 1994. A manual for primary animal health care worker. FAO Animal Health Manual. http://www.fao.org/docrep/t0690e/t0690e09.htm.
Foster, M.L. and S.R. Humphrey, 1995. Use of Highway underpasses by florida panthers and other wildlife. Wildlife Soc. Bull., 23: 95-100.
Direct Link |
GMS, 2002. General motors corporation and delphi-delco electronic systems. Automotive Collision Avoidance Operational Field Test Warning Cue Implementations Summary Report. Publication No. DOT-HS-809-462, National Highway Transportation Safety Administration, U.S. Department of Transportation, 2002.
Gagnon, J.W., N.L. Dodd, S. Boe and R.E. Schweinsburg, 2009. Using global positioning system technology to determine wildlife crossing structure placement and evaluating their success in Arizona. International Conference on Ecology and Transportation, Sept. 2009.
Gagnon, J.W., T.C. Theimer, N.L. Dodd, S. Boe and R.E. Schweinsburg, 2007. Traffic volume alters elk distribution and highway crossings in Arizona. J. Wildlife Manage., 71: 2318-2323.
CrossRef | Direct Link |
Gordon, K.M. and S.H. Anderson, 2001. Motorist response to a deer-sensing warning system in western Wyoming. Proceedings of the 2001 International Conference on Ecology and Transportation, September 24-28, 2001, Raleigh, NC., pp: 549-558.
Gordon, K.M., 2001. Evaluation of the FLASH (Flashing Light Animal Sensing Host) System in Nugget Canyon, Wyoming. Wyoming Dept. of Transportation, Laramie, WY, USA..
Graham, R., 1999. Use of auditory icons as emergency warnings: Evaluation within a vehicle collision avoidance application. Ergonomics, 42: 1233-1248.
Haikonen, H. and H. Summala, 2001. Deer-vehicle crashes: Extensive peak at 1 hour after sunset. Am. J. Prev. Med., 21: 209-213.
Harris, C.C., 2005. Creature of the desert, camel. http://www.touregypt.net/ featurestories/camel.htm.
Hedlund, J.H., P.D. Curtis, G. Curtis and A.F. Williams, 2004. Methods to reduce traffic crashes involving deer: What works and what does not. Traffic Inj. Prev., 5: 122-131.
Hirota, M., Y. Nakajima, M. Saito and M. Uchiyama, 2004. Low-cost infrared imaging sensors for automotive applications. Adv. Microsyst. Automotive Appl., 2004: 63-84.
Honda Motor Co. Ltd., 2004. Intelligent night vision system able to detect pedestrians and provide drivers cautions. Automobile News, August 24, 2004. Honda Motor Co. Ltd., Japan.
Huijser, M.P. and P.T. McGowen, 2003. Overview of animal detection and animal warning systems in North America and Europe. Proceedings of the 2003 International Conference on Ecology and Transportation, (ET'03), Raleigh, NC, USA., pp: 368-382.
IRD, 2002. Wildlife warning system. IRD (International Road Dynamics), Saskatoon, SK, Canada.
Kais, M., L. Bouraoui, S. Morin, A. Porterie and M. Parent, 2005. A collaborative perception framework for intelligent transportation systems applications. Proceedings of the 12th World Congress on Intelligence Transport System, Nov. 6-10, San Francisco California, USA., pp: 35-42.
Kistler, R., 1998. Wissenschaftliche begleitung der wildwarnanlagen calstrom WWA-12-S. July 1995-November 1997. Schlussbericht. Infodienst Wildbiologie and Oekologie, Zurich, Switzerland.
Langley, R.L., S.A. Higgins and K.B. Herrin, 2006. Risk factors associated with fatal animal-vehicle collisions in the United States, 1995-2004. Wilderness Environ. Med., 17: 229-239.
Lee, J.D. and J.D. Hoffman, 2004. Collision warning designed to mitigate driver distraction. Presented at SIGCHI Conference on Human Factors in Computing Systems, Vienna Austria.
Luo, J. and J.P. Hubaux, 2004. A survey of inter-vehicle communication. School of Computer Communication Science., Swiss Federal Institute of Technology., Lausanne, Switzerland.
Macklean, G., 2009. Weak GPS signal detection in animal tracking. J. Navigation, 62: 1-21.
Maltz, M., H. Sun, Q. Wu and R. Mourant, 2004. Use of in-vehicle alerting system for older and younger drivers: Does experience count. J. Transport. Res. Board, 1899: 64-70.
McKinney, T. and T. Smith, 2007. US93 bighorn sheep study: Distribution and trans-highway movements of desert bighorn sheep in Northwestern Arizona. Final Report to Arizona Department of Transportation, JPA04-032T/KR04-0104TRN.
National Geograhic, 2009. Arabian (dromedary) camel. http://animals.nationalgeographic.com/animals/printable/dromedary-camel.html.
Orazem, V., 1995. Holography as an element of the media architecture. Proc. Int. Symp. Display Holography, 2333: 168-177.
Pojar, T.M., R.A. Prosence, D.F. Reed and R.H. Woodward, 1975. Effectiveness of alighted, animated deer crossing sign. J. Wildlife Manage., 39: 87-91.
Direct Link |
Putman, R.J., 1997. Deer and road traffic accidents: Options for management. J. Environ. Manage., 51: 43-57.
CrossRef | Direct Link |
Pynn, T.P. and B.R. Pynn, 2004. Moose and other large animal wildlife vehicle collisions: Implications for prevention and emergency care. J. Emerg. Nurs., 30: 542-547.
Ragab, K., N. Kaji and K. Mori, 2003. Service oriented autonomous decentralized community communication technique for a complex adaptive information system. Proceedings of the IEEE/WIC International Conference on Web Intelligence, (WI`03), Halifax, Canada, pp: 323-323.
Ragab, K., N. Kaji and K. Mori, 2004. ACIS: A large scale autonomous decentralized community communication infrastructure. IEICE Trans. Inform. Syst., E87: 937-946.
Ragab, K., N. Kaji, H. Kuriyama and K. Mori, 2003. Scalable multilateral communication technique for large-scale information systems. Proceedings of the 27th Annual International Conference on Computer Software and Applications, (PSC`03), USA., pp: 222-222.
Ragab, K., N. Kaji, H. Kuriyama and K. Mori, 2004. Autonomous decentralized community communication for information dissemination. J. IEEE Internet Comput., 8: 29-36.
Ragab, K., T. Ono, N. Kaji and K. Mori, 2002. Community communication technology for achieving timeless in autonomous decentralized community systems. The 2nd International Workshop on IWADS, (IWADS`02), Beijing, China, pp: 56-60.
Reed, D.F., 1981. Effectiveness of highway lighting in reducing deer-vehicle collisions. J. Wildlife Manage., 45: 721-726.
Direct Link |
Reed, D.F., T.D.I. Beck and T.N. Woodard, 1982. Methods of reducing deer-vehicle accidents: Benefit-cost analysis. Wildlife Soc. Bull., 10: 349-354.
Direct Link |
Romin, L.A. and J.A. Bissonette, 1996. Deer-vehicle collisions: Status of state monitoring activities and mitigation efforts. Wildlife Soc. Bull., 24: 276-283.
Direct Link |
Rutter, S.M., 2007. The integration of GPS, vegetation mapping and GIS in ecological and behavioural studies. R. Bras. Zootec., 36: 63-70.
STS, 2002. Roadway animal detection system. Operator's Manual and Installation Guide. RADS-00-007-OM, 16 October 2002, Revision 4. Scottsdale, Arizona, USA.
Shipley, L.A., 2001. Evaluating Wolfi n as a repellent to wildlife on roads in Washington and the feasibility of using deer-activated warning signs to reduce deer- automobile collisions on highways in Washington. Research Report Research Project No. T9902. Department of Natural Resource Sciences, Washington State University, Pullman, WA, USA.
Stout, R.J., R.C. Stedman, D.J. Decker and B.A. Knuth, 1993. Perceptions of risk from deer-related vehicle accidents: Implications for public preferences for deer herd size. Wildlife Soc. Bull., 21: 237-249.
Direct Link |
Sullivan, T.L., A.F. Williams, T.A. Messmer, L.A. Hellinga and S.Y. Kyrychenko, 2004. Effectiveness of temporary warning signs in reducing deer-vehicle collisions during mule deer migrations. Wildlife Soc. Bull., 32: 907-915.
Direct Link |
Tan, H.S. and J. Huang, 2006. DGPS-based vehicle-to-vehicle cooperative collision warning: Engineering feasibility viewpoints. IEEE Transportation Intell. Syst., 7: 415-428.
Direct Link |
Tijerina, L., F.S. Barickman, M.D. Winterbottom, H.A. Pham, E. Parmer and S. Johnston, 2000. Preliminary Studies in Haptic Displays for Rear-end Collision Avoidance System and Adaptive Cruise Control Application. National Highway Transportation Safety Administration, Washington, D.C., pp: 159.
Turner, L.W., M. Anderson, B.T. Larson and M.C. Udal, 2001. Global Positioning Systems (GPS) and grazing behavior in cattle. Proceedings of the 6th International Symposium on Livestock Environment, May 21-23, 2001, Louisville, KY. USA., pp: 640-650.
Ward, A.L., 1982. Mule deer behavior in relation to fencing and underpasses on Interstate 80 in Wyoming. Transp. Res. Rec., 859: 8-13.