Bạn đang xem bản rút gọn của tài liệu. Xem và tải ngay bản đầy đủ của tài liệu tại đây (224.68 KB, 22 trang )
<span class="text_page_counter">Trang 1</span><div class="page_container" data-page="1">
HANOI UNIVERSITY OF SCIENCE AND TECHNOLOGYSCHOOL OF ELECTRICAL AND ELECTRONIC ENGINEERING
The development of smart cities and the rising need for the Internet of things (IoT)have made the accident management and detection system a common research topic.The increased vehicle congestion on the routes has made it more difficult to handleaccidents in the centers. The primary factor in human life loss is the failure to transferpatients and provide first aid and medical care when necessary. This paper explores thekey opportunities brought about by the introduction of the concepts of IoT network. Theidea of a vehicular network and the Internet of things is intended to give emergencyvehicles priority. The suggested approach aids in accident detection and emergencymedical assistance at an appropriate time.
I would like to thank Ms. Bui Van Anh for the template and Dr. Nguyen Tien Hoafor his constant support and inspiration during the process of preparation of this paper.
</div><span class="text_page_counter">Trang 3</span><div class="page_container" data-page="3">3.1 Algorithm . . . . 63.2 Performance parameters . . . .
4.1 Simulation situation . . . . 84.2 Result . . . . 8
</div><span class="text_page_counter">Trang 4</span><div class="page_container" data-page="4">Figure 2.1 System general blocks . . . . 4
Figure 2.2 Accident management . . . . 5
Figure 3.1 System algorithm . . . . 6
Figure 4.1 IoT scenario . . . . 8
Figure 4.2 VANET scenario . . . . 9
Figure 4.3 Comparison of routing protocol on the basis of Average Throughput 9Figure 4.4 Energy consumed by sensors . . . . 10
Figure 4.5 Remaining energy . . . . 10
Figure 4.6 Transmitting energy . . . . 11
Figure 4.7 Average throughput in case of VANET . . . . 12
</div><span class="text_page_counter">Trang 6</span><div class="page_container" data-page="6">Table 3.1 IoT simulation Parameters . . . .Table 3.2 VANET simulation Parameters . . . . 7
</div><span class="text_page_counter">Trang 7</span><div class="page_container" data-page="7">Death and disabilities by road accident are increasing every passing year. In manycases, human lives are lost due to delays in emergency medical assistance. The rising ofvehicle production has helped the large cities’ traffic, however the increase in traffic onthe roads prevents emergency vehicles from getting where they need to go.
In the recent past, information and communication technology such as IoT has beenused to decrease the accident rescue time. IoT is an interconnection of vast variety ofembedded and smart devices such as computers, smartphones, smart sensors and actu-ators. IoT is a potential medium for tracking and control of smart automobiles that canlink any connected physical unit to a control server. Most researchers have confinedtheir work to improving the accuracy of accident detection, estimating the severity ofroad accidents or minimizing the rescue time postoccurrence of an accident. In terms ofdrawbacks, most systems to detect and report road accidents are expensive and limitedto high-end vehicles. Another drawback of the current systems is their inability in iden-tifying the type of accident as a collision, rollover or a fall-off event. Merely reportingdata of an accident event severely limits the ability of the emergency rescue workers toprovide the victims with the right kind of rescue support and medical aid.
This research work answers the following questions:
1. Can there be an inexpensive accident detection and classification system that can beattached to any vehicles?
2. Which is the best-suited protocol model that can accurately detect and classify theroad accident type while consuming less energy?
3. How to automatically report the occurrence of road accident with its scale and tion by sending the emergency notification if the accident victims are incapacitated?
</div><span class="text_page_counter">Trang 8</span><div class="page_container" data-page="8">The amount of traffic in the modern day is getting progressively worse. Accordingto estimates, by the year 2035, there will be more than 1.2 billion vehicles [1], which willcause oppressive levels of traffic. Vehicle production is rising quickly. This has helpedthe cities’ traffic congestion, which frequently results in serious collisions. Additionally,the increase in traffic on the roads prevents emergency vehicles from getting where theyneed to go, which is the main reason for the loss of precious lives. This has been a majorchallenge for authorities concerning road traffic management and accident detection,therefore, continues to be a significant problem that has to be resolved. In this regard,the VANET (Vehicular - adhoc - network) have attracted considerable attention from theresearchers.
An IoT (Internet - of - Things) sensor system is built into a vehicle unit to helpit identify accidents and alert a roadside unit when one occurs. All warning messagesare received by a roadside unit, which relays them to the rescue crew. By implementingVANET, automobiles can create effective communication with one another, transmittingsafety messages and different cautions like lane change warnings and emergency vehiclearrival notifications among others. Some specific opportunities and constraints needto be considered in order to enable these technologies. It is suggested to use a smartarchitecture for emergency rescue services that incorporates all of these technologies.The use of VANET to identify an accident and it offers crucial information: passengersrequire medical aid, and an accident has occurred on a large or small scale [2]. IoTdevices are utilized for establishing a link between the devices, then transmit data to thecenter unit.
To achieve the goal, the IEEE 802.11p standard is in charge of setting up a work in automobiles. An industry standard for short-range communication called Waveprotocol encourages the vehicle to the roadside unit and directs the device to devisecommunication.
net-This paper aims to establish an accident management system that detects the sion and helps to provide medical assistance. Sensor devices are deployed in automobilesto identify the scenario and accident involving a passenger [3]. The on-board unit andtraffic camera send information about the accident to the processing unit. After the im-age detection, it is confirmed that the accident happened. The affected area is computedbased on the information collected about the location and disseminate the alert messageto the specific dashboard and on-board units. The objective of this process is to detectthe accident and provide the medical assistance on the accident location.
</div><span class="text_page_counter">Trang 9</span><div class="page_container" data-page="9">colli-The research project can be divided into the following two features. In Chapter 1,this article summarizes related work in the area of accident management systems. Fur-thermore, the architecture for accident management system is designed and each compo-nent’s functionality is assessed. Similarly, the communication between various segmentsis examined with the aid of connotative technologies. Thanks to the help of VANET andan IoT simulation tool (Netsim), the suggested architecture is expanded on a simulatedscenario, and the performance is then determined. Chapter 2 gives the modern status ofstudy in the fields of an accident management system. The architecture design for theaccident management system is then illustrated in Chapter 3. Finally a semiotic modelfor collision control systems is introduced in Chapter 4. These chapters emphasize theimplementation of the specified architecture using a simulation model and scenario.
</div><span class="text_page_counter">Trang 10</span><div class="page_container" data-page="10">There are several researchers’ utilized distinct methods to analyze accident events,where accident management is a well-studied area. Also, particular kinds of automaticsensors applied to identify accident events and smart-phones based applications are usedto interact among the sensors by utilizing portable data networks as the transmissionmedium. A number of clarifications stated in the research are given in this chapter.
2.1 Network protocols
Collision management is a well-researched area where different approaches are ommended for collision detection warning. Moreover it also contains biometric sensorsfor temperature and heart rate that are used to identify accidents.
rec-Emergency responders are positioned and informed about conflicting vehicles ing GPS (Global Positioning System) and GSM (Global System for Mobile communi-cation) [4]. Smartphone’s microphone, gyroscope, and accelerometer can be utilizedto recognize collisions. This approach generates alerts throughout GPS connections.Vehicle sensors such as airbag, accelerometer, and angle sensors can detect collisionsand apply upcoming IP interface (networks) to broadcast the warning [5]. The primaryclaims for crossing circumstances are addressed, which also uses visual processing tech-niques to observe the carrier on the vehicular network. To request remedical treatmentfrom the nearby hospitals, a message is transmitted using an Android based application[6]. The route planning that is based on GPS is conducted with the aid of the Dijkstraalgorithm. However, the major shortcoming in this method is the deficiency to plan aroute [7].
us-An effective path preparation approach that is based on the VANET and optimalpath is explored by implementing A* algorithm [8]. Various methods are used to de-termine the optimal route by utilizing the (Stochastic Lyapunov) optimization method[9]. An approach proposed to prioritizing traffic light by using wireless devices and itadditionally develops the intensity of the alarm for the emergency carrier to recognizedirection and length for the emergency vehicles [10]. An IoT and RFID-based (Ra-dio Frequency Identification) vehicle identification and data collection system is usedto control and monitor the vehicles. An architectural design for distributing accidentwarning messages to prevent further collisions [11].
2.2 VANET network
The major objective of this research is to come up with a solution that can fill outrequirements while decreases the cost of deployment. VANET is the most beneficial
</div><span class="text_page_counter">Trang 11</span><div class="page_container" data-page="11">answer to the critical obstacle since modern cars are implemented with wireless faces. Rescue services are provided using the communication technologies V2I (vehicleto service unit) and V2V (communication between vehicles) [6]. A vehicle ad hoc net-work, or VANET, is designed to operate in a vehicular network environment at a lowdeployment charge. This technology is applied to disseminate the message in the vehic-ular environment and transmit alert to the neighboring vehicles to report the collisionsand utility of another path, so an ambulance will take the less congested path to providerescue services in a collision situation.
inter-2.3 Architecture design
Many frameworks for accident management systems have already been developedby researchers, however no research has focused on creating a specific architecture to-wards an accident management system. This segment illustrates the key design elementsthat represent the architecture designing for such systems. As shown in Fig 2.1, thereare several major units including onboard unit, traffic camera, accident alert unit anddashboard.
Figure 2.1 System general blocks
• On-board unit (OBU) is in charge of the movements conducted in a vehicle thatincludes collision discovery and interaction with other vehicles. It includes various
</div><span class="text_page_counter">Trang 12</span><div class="page_container" data-page="12">happened. The affected area is computed based on the information about the cation and alert is disseminating to the appropriate on-board unit and dashboard.This module aids in discovering the collision, and data is gathered via the on-boardunit’s information system.
lo-• Accident alert unit and dashboard While the collision is identified, it nates a warning message to the server room in a specific area by utilizing Vehicu-lar communication (V2V). In order to request medical assistance (ambulance), theserver room collects information about the accident and sends a warning messageto the hospital.This architecture is later modeled using 20 vehicles. Depicted inFig 2.2 ambulance also establishes V2V Communication with the vehicles to reachthe hospital without any delay by following the procedure. Initially, the ambulanceprovides the active warning to drive left as it approaches. To avoid a collision, thevehicle in the opposite path receives BSM (Blindspot Monitoring) from other ve-hicles. A second car is then given the information and is instructed to drive at amoderate speed.
dissemi-In general the straight path in Fig 2.2 is suggested in order to avoid traffic and savetime. The vehicles provide information to others to drive slowly and allow an ambulanceto pass. The message is then transmitted to the next vehicle to follow the straight path.The process results in the straight lane of vehicles that provides the information allowingthe ambulance to pass when it approaches the hospital.
Figure 2.2 Accident management
</div><span class="text_page_counter">Trang 13</span><div class="page_container" data-page="13">3.1 Algorithm
Netsim is employed in this research to carry out the simulation. The scenario forIoT is illustrated using wireless sensors, wired nodes and low pan gateways. The mapis exported from the OSM (Open Street Map) to Netsim and the desired scenario ismodeled using vehicles, ad-hoc link, and wireless links. The simulation methodologyfor IoT and VANET scenario is shown in Fig 3.1
</div><span class="text_page_counter">Trang 14</span><div class="page_container" data-page="14">The routers are utilised the routing protocols to determine the optimal way forsuccessfully information distribution between sensors. The average throughput is variedwith respect to RPL (Routing Protocol for Low power networks) and AODV (Ad-hoc Onthe Demand Distance Vector) routing protocol, specified in Table 3.2.
• AODV is a route tracking process. This mechanism is begun after it transmits thedata and RREQ (Route Request) packets to the neighbor nodes [13].
• RPL is a proactive routing protocol which means it has a moderate energy sumption. It is also very responsive to packet loss for wireless networks due to theability of sharing efficient knowledge about the routing within the network. Thisprotocol supports a 1-to-1 messages system [14].
Mobility model Random way point, Random walk
Table 3.2 VANET Simulation Parameters
The path for mobile devices is assessed using the random waypoint mobility model.It is used to determine the acceleration, position, and velocity. The moveable objectsmove impulsively and without any restrictions.
The random walk is a scientific method describing a route that consists of somescientific values. Observed performances of several routing activities are represented byrandom walks.
</div><span class="text_page_counter">Trang 15</span><div class="page_container" data-page="15">4.1 Simulation situation
The wireless sensorApp1 serves as an on-board device, continuously monitoringthe orientation and acceleration of the vehicle in order to identify collisions in the sce-nario. After a crash is discovered, the most important information, including the vehi-cle’s location and temperature, is stored in the storage node.
Figure 4.1 IoT scenario
As shown in Fig 4.1, the OBU gathers the information regarding the collided cle and transfers it to the wired node. Traffic cameras, which are modeled by the wirelesssensorApp2, wireless sensorApp3 in this research, are in charge of collecting the imagesfrom the accident location and transfer the information to the wired node. The affectedarea is later computed by the wired node based on the data about the location and alert
</div>