Keywords

1 Introduction

Billions of people are losing their lives due to accidents. The accidents are the most uninvited, unintentional event that causes lot of damage and even cost person’s life. Road accidents are the major integral part of accidents. Road accidents occur due to collision of vehicle with another vehicle or any object. Road accidents are responsible for almost 2.2 million of deaths all over the world each year. Death by accidents has increased to a greater level and has become the second largest reason for death after heart disease. Road injuries have resulted in death of many people in 1991; it was around 1.0 billion which has been increased up to 1.8 billion in 2013. Major reason for road accidents is driving at excess speeds, poorly under-developed highways, poor traffic rules, poor infrastructure and unattended zones. Road accidents occur daily in large number. Most prone areas are highways or sharp turns over road ends. By integrating VANETs with sensor networks, road accidents can be reduced to a greater level. The sensors will send their vehicle’s information to the other vehicle using ZigBee. The exchange of information will occur only when the two ZigBee embedded vehicles will fall within the range. Also the position of the nearby vehicle will be displayed on an android application. Figure 1(a) below shows the death rate over years. Figure 1(b) shows the fatality rate due to road accidents in various states of India in past two years.

Fig. 1.
figure 1

(a) Traffic death per year (b) Fatality rate due to road accidents in various states in year 2014 and 2015

Full size image

2 Literature Review

See Table 1.

Table 1. Literature review
Full size table

3 Area of Concern

In this paper, a method is proposed to reduce number of collisions to reduce death rates. Collisions likely occur at blind turns, on highways during lane changing and on hilly areas. So an effort has been made to avoid accidents in these areas so as to make world a better and safe place to live Table 2.

Table 2. Various techniques along with their problems
Full size table

4 Proposed Work

This system works for accident prone areas like highways, dead ends and blind turns. The selection of area is the most crucial part of the implementation of the proposed work. The vehicle moving in the areas will act as sensor nodes. Users must have installed android application (developed in this work) on their mobile devices. This application will also help to track the other vehicles over Google Map. For the development of proposed model, the work has been divided into two phases. The first phase involves implementation of prototype and the next phase is to develop android based application for generation of alert signals and to trace location of vehicles on roads.

DETECTION stage:

This stage first checks the vehicle is either standing or in motion. If standing, check if it is a blind turn (i.e. vehicle making angle > 90° or < 270° with the road), then collision is likely to occur. If the vehicle is moving, check if the approaching vehicle is within ZigBee’s range. It is assumed that these vehicles have ZigBees installed in them and whenever a vehicle comes within this range, an alert message will be generated. The same scenario can be considered for vehicles coming from front as well as from back.

ALERT stage:

An alert in the form of buzzer beep will be generated and a message will be flashed on the LCD screen installed inside the vehicle. After the driver gets alert, he then checks the android application to get the location of nearby heading vehicle on Google map. After getting the location, the driver acts accordingly to avoid collision. An algorithm for the scenario is defined below.

figure a

5 Design Philosophy

5.1 Hardware Module

To verify the working of the proposed system, a hardware prototype is developed. A block diagram showing hardware module representing vehicle is shown in Fig. 2. In this, IR sensor is used as a proximity sensor as these sensors have minimum range. A ZigBee attach with the vehicle is used for communication purpose. This protocol has a pre-defined range which is 70–80 m. When a vehicle falls within the ZigBee’s range of another vehicle, an alert warning in the form of buzzer will be generated and a message will be flashed on the LCD screen equipped inside the vehicle. Crystal in the model is used to avoid oscillations in the power supply. Microcontroller works at 5 V, the power supply that we are providing is via adapter or by 9 V battery.

Fig. 2.
figure 2

Block diagram of a vehicle

Full size image

Figure 3 given below shows the circuit diagram of a vehicle in detail:

Fig. 3.
figure 3

Circuit diagram of a vehicle

Full size image

  • LED is connected with 5 V VCC along with a resistor, VCC is also connected to one end of 100 microFarad capacitor whose other end is grounded.

  • There is an LCD whose pin VSS, RW and K is grounded, VDD is connected to 5 V VCC, VO is connected to a resistance of 10 k along with the load of 1 k, RS is connected to PB5 of microcontroller, E is connected to pin PB4 of microcontroller, DB4, DB5, DB6 and DB7 is connected to pin PD4, PD5, PD6, and PD7 of microcontroller respectively and A is connected to 5 V VCC along with the load of 1 k.

  • A Tachometer has 3 pins out of which one is grounded, other one connected to pin PD2 of the microcontroller and the third one is connected to 5 V VCC. ZigBee has four pins one of which is ground, other is connected to 5 V VCC and rest two are input and output pins connected to PD0 and PD1 of microcontroller respectively.

  • An RF Module has four pins connected to ground, 5 V VCC, ANT and TH12D respectively. TH12 takes one connection from RF Module and split it into two lines connected with pins PC0 and PC1 which are AC-to-Dc converter. A microcontroller having 28 pins.

A prototype presenting the circuit of the vehicle is shown in Fig. 4(a). The circuit is mounted over vehicle prototype whose movement is controlled via remote control. The prototype is designed to note how vehicle prototype will behave during the emergency, before actually implementing it on real scenarios.

Fig. 4.
figure 4

(a) Vehicle prototype (b) Remote control

Full size image

The remote control is shown in Fig. 4(b). IR sensors are used for sending and receiving signals. These sensors control the vehicle movement in either forward or backward direction.

5.2 Android Application

Android application to track the approaching vehicle was developed. This application is executed in phases.

  • Phase1: In this phase the application will get the current vehicle’s location via GPS on the display screen of the cellular device.

  • Phase2: The location captured is then transfer to the server where it is stored for later use. This transfer is done with the help of File Transfer Protocol.

  • Phase3: From server this application fetches location of all the other vehicles except for its own location. This process also takes place via File Transfer Protocol.

  • Phase4: All the location accessed by the application i.e. its own along with other vehicles are plotted over Google Map with different color markings(purple for its own and cyan for other vehicles) as shown in Fig. 6.

Figure 5 shows the four phases of android application.

Fig. 5.
figure 5

Four phases of Android application

Full size image

Android application code consists of five classes.

  • GPSTracker class make use of Location Manager which provide the location of vehicle.

  • Background Task class take the url of the server and POST the data onto the server.

  • Track class basically takes the location and separates them with a comma and then returns the values.

  • Maps Activity class takes latitude and longitude values from server and plot it into Google Map.

  • Main Activity class which is the main class, it makes use of Telephony Manager which is for getting the id of Mobile phone. This class calls all the above classes. Figure 6 below shows the screenshot of the android application.

    Fig. 6.
    figure 6

    Android application screenshots

    Full size image

6 Results

For evaluation of the above proposed model, different scenarios are considered. The results have been analyzed on monitoring the behavior of the vehicles.

Scenario 1:

Vehicles within range of each other

When approaching vehicles comes within each other range, a buzzer will buzz and a message will be displayed on LCD screen “Vehicle Detected Nearby”. Vehicle position can be seen on an android application installed on user’s mobile. User can clearly see the position of approaching vehicle and will get sufficient time to take suitable action.

Scenario 2:

Vehicle moving infront of other vehicle

When there is a vehicle in front of another vehicle. At some specific range a buzzer will ring to alert the driver and also a message will be displayed on LCD screen as “Vehicle in the Front Wake Up”.

Scenario 3:

Vehicle moving at the back of other vehicle

An IR sensor is also mounted at the back of the vehicle to alert the driver about approaching danger from back. A buzzer alert along the message on LCD “Vehicle at the Back Beware” will be signaled.

7 Conclusion and Future Work

The vehicle prototype mentioned in the paper is a defensive way of approaching the problem of road accidents by vehicle collision. Road accidents will be reduced by avoiding the collision of vehicles. There will be a buzzer alert inside the vehicle whenever the vehicle falls within the ZigBee range, a message will be displayed on the LCD of the vehicle stating “Vehicle Detected Nearby”, the driver then will check the android application which will give the exact location of the preceding vehicle. In this way collision between the vehicles will be avoided. In addition to this, if there is a vehicle preceding from the back on a great speed the proximity sensor at the back of the vehicle will detect it and there will be an alert message in the front car stating “Vehicle Detected at the Back” and in the back car will be “Vehicle Detected in the Front Wake Up”. Further if this prototype is made to work upon real time in future there will be changes in the type of sensors used, as then the sensors having large range will be used and the environment conditions will also be taken into account.